**** BEGIN LOGGING AT Wed Jan 22 02:59:59 2014 Jan 22 03:13:47 hmm, i set the imeActionLabel, but this doens't appear to be reflected in the keyboard Jan 22 03:17:36 figured it out, i had to set an inputtype for it to work Jan 22 03:45:06 installed haxm and the atom image, but when I try to create a new AVD the cpu always goes to ARM and is grayed out Jan 22 03:45:59 the atom image directory is right there next to the arm one Jan 22 03:48:02 I just have api level 19 Jan 22 03:48:12 so by default, it seems like my edittext will hide if i press the enter action Jan 22 03:48:21 is there some way to stop the keyboard from hiding Jan 22 03:51:17 nevermind, restarted eclipse Jan 22 03:53:00 nice. faster Jan 22 03:57:30 so i have eclipse helios, is there any reason to update? Jan 22 03:57:59 did a new eclipse come out ? Jan 22 03:58:35 Kepler? Jan 22 03:59:40 the ADT uses 3.7 workbench apis, doesn't matter much Jan 22 03:59:45 kepler has been out for quite a long time Jan 22 04:00:01 i thought 3.8.1 was the best eclipse ;) Jan 22 04:00:06 my eclipse probably is ancient Jan 22 04:00:06 same version I used in college lol Jan 22 04:01:08 I use 3.7.1 Jan 22 04:01:08 bit more up to date than college one but still bit old lol Jan 22 04:01:23 numeral: Yeah, I meant 'isn't it still kepler?' :) Jan 22 04:01:49 Node: yeah that makes much more sense Jan 22 04:10:57 huh, when I inflate a layout, shouldn't layout params be immediately available? Jan 22 04:11:12 or do they only apply if I specify a parent container... Jan 22 04:16:36 pfn: does this help? http://www.doubleencore.com/2013/05/layout-inflation-as-intended/ Jan 22 04:17:33 anyway, layoutparams are only populated if specifying a parent to inflate Jan 22 04:18:56 kinda sounds familiar Jan 22 04:23:23 layout params originate from the parent view group's generateDefaultLayoutParams method Jan 22 04:24:30 yeah, figured as much Jan 22 04:26:11 thanks Jan 22 04:29:02 hi, android lint when looking for errors takes forever, restart eclipse did not wake up lint. what i should do to make it work? Jan 22 04:31:29 hoo, lint came back, damn slow Jan 22 04:38:30 dragorn http://blog.cyberexplorer.me/2014/01/sniffing-and-decoding-nrf24l01-and.html Jan 22 04:38:46 g00s: I saw that Jan 22 04:39:38 the downconverter is the magic Jan 22 04:51:51 hello, where i can find code example of a soft back button implementation? Jan 22 04:54:39 what do you mean ? Jan 22 04:55:24 <_genuser_> interesting, google still hasn't processed payment for dev account for play store. Jan 22 04:55:39 bankai_: i have a button working as back to previous list. i need to change its shape. just like a back button has lef arrow head Jan 22 04:55:51 <_genuser_> make payment yesterday morning CST. they charged 1 dollar then reverse. They haven't charged 25 yet. Jan 22 04:57:22 <_genuser_> anybody else have delayed registration? Jan 22 04:57:40 bankai_: how to do that Jan 22 04:59:51 any idea? Jan 22 05:00:27 i still don't know what you're asking, you have a back button on the action bar you want to style ? Jan 22 05:01:40 no, i have a button working as back to listview , but the button is in rectangular shape, i want to change shape of button so it look like soft back button Jan 22 05:01:48 is there nothing I can do to make the ContextMenu that pops up on long press , black background ???? Jan 22 05:02:29 <= like that Jan 22 05:03:12 bankai_: my question is still not clear? Jan 22 05:03:57 :) Jan 22 05:04:10 look like i need to pay google for elp Jan 22 05:04:15 help** Jan 22 05:04:30 jaami-win: so change it, what's the problem ? Jan 22 05:04:39 lol Jan 22 05:04:42 get the image you want, set it, then go Jan 22 05:04:54 i dont know how to change that is why asking you Jan 22 05:05:11 oops, is that so Jan 22 05:05:23 easy as setting background? Jan 22 05:05:26 setBackgroundResource or use an imageview or something Jan 22 05:05:42 err ImageButton* Jan 22 05:05:59 <_genuser_> if you're loading something in action bar, you should be loading it from a menu that gets inflated. Jan 22 05:06:12 ahh okay, there is the tip... thanks a lot. i guess that is what i needed to know Jan 22 05:06:40 <_genuser_> android:icon="@drawable/name_of_your_resource" Jan 22 05:07:00 <_genuser_> when you define your menu, use the attribute above in your menu item tag and load a resource from drawables. Jan 22 05:07:49 _genuser_: those wouldn't work for popup ContextMenu would they ? Jan 22 05:08:18 <_genuser_> shmooz: pop menu? i.e. when you hit the meny button? Jan 22 05:08:22 I know its someone elses answer but I could use setBackgroundResource on ContextMenu if it worked Jan 22 05:08:37 when I long press the popup menu Jan 22 05:09:06 I mean when I long press the list item, the background and text color of the ContextMenu that pops up Jan 22 05:09:14 _genuser_: & bankai_ thanks , i go try that Jan 22 05:09:24 <_genuser_> shmooz: loading an image should work on context menus. I say should, but I don't know, I haven't tried it. Jan 22 05:09:50 otherwise I'll have to make my own custom menu Jan 22 05:09:56 <_genuser_> shmooz: from the documentation it seems to be just a menu that gets inflated in onCreateContextMenu(). Jan 22 05:10:19 _genuser_: yeah that's what I did Jan 22 05:10:31 but it's godawful white background Jan 22 05:10:54 <_genuser_> try android:background="#000000" and see if that changes it. Jan 22 05:10:59 and I tried setting global theme to override too Jan 22 05:11:18 _genuser_: I tried that too, it had no effect Jan 22 05:11:18 <_genuser_> it should. And if it works, you can set a drawable resource, with gradients, blah blah. Jan 22 05:11:38 ok maybe I missed something, I'll try again Jan 22 05:11:41 <_genuser_> I see. then I'm afraid we've reached the end of what I know about context menus. :) Jan 22 05:12:16 same problem with the dropdown spinner, I was able to set the background color from xml, but I couldn't find out how to set the text color Jan 22 05:12:34 so the spinner shows black on black for now Jan 22 05:12:34 <_genuser_> android:textColor="" Jan 22 05:12:51 _genuser_: tried that, didn't work either Jan 22 05:13:31 those properties work for most widgets, but ContextMenu and Spinner, it seems to have no effect Jan 22 05:14:03 <_genuser_> http://stackoverflow.com/questions/2944244/change-the-background-color-of-the-options-menu Jan 22 05:14:11 <_genuser_> would seem to indicate it's a long observed problem. Jan 22 05:15:18 _genuser_: thanks for the help, I did see that page before too, but I'll read more carefully to see if I missed something Jan 22 05:15:34 oh well I can always popup a custom menu right Jan 22 05:15:36 <_genuser_> shmooz: somebody mentions android:panelFullBackground below in the response. try that one. Jan 22 05:15:49 _genuser_: ok Jan 22 05:19:23 _genuser_: the problem with that is I won't be able to set the text color to white, but oh well at least I will be able to make it black on gray background if it works Jan 22 05:19:44 anything is better than white Jan 22 05:19:49 <_genuser_> lol, it really is a freak of a problem. Jan 22 05:50:28 what's the problem? Jan 22 05:53:18 Hey Guys, I was looking for introducing automation testing for my mobile app Jan 22 05:53:43 but I find writing test cases difficult as code gets spread across Activities, Services AsyncTasks Jan 22 05:54:15 Is there a good project I could look at, to learn how to do automation testing for an android app Jan 22 06:05:15 ~~~~~/close Jan 22 06:05:55 JakeWharton: the problem is there seems to be no easy way to set background and text colors on the ContextMenu that pops up after long pressing a list item Jan 22 06:07:12 ah, i thought you guys were talking about action bar popups like spinner or overflow Jan 22 06:07:56 well I do have a similar problem on Spinner, I can set background color fine, but couldn't change the text to white Jan 22 06:08:21 the text color is based on the layout you use in the adapter Jan 22 06:08:31 and the context that's used to inflate it Jan 22 07:19:24 hi Jan 22 07:19:30 i am storing something as a blob Jan 22 07:19:40 but what if I have nothing in the byte array? Jan 22 07:19:56 when I pass an object where the byte array is not initialised through an intent, I get a null pointer exception Jan 22 07:20:09 but if I initialise the array, say x = new byte[0]; Jan 22 07:20:45 then how can I tell on the next activity whether the byte array has real data in it (ie. image data) or whether it's just my fake initialisation to avoid the NPE? Jan 22 07:20:56 checking the length seems like a crude way to do it Jan 22 07:21:02 is there a better alternative Jan 22 07:25:36 is there some way to interrupt/prevent a softkeyboard's action from hiding the keyboard Jan 22 08:33:07 can someone tell me his experience on using ant/maven/gradle for android development? i have done a little ant today, but since i'm used to developing JEE projects with maven, it seems like an awful lot of typing and double checking. i know almost nothing about gradle except that it's almost finished and already usable and supposed to be the future of android development (because it's being pushed/developed by google). is it wo Jan 22 08:33:08 rth checking out yet? Jan 22 08:40:42 Yes, use gradle for Android stuff. Although I know nothing about the current state of maven support for Android projects. Jan 22 08:41:01 Is android studio stable? Jan 22 08:41:42 Is it worth switching over from eclipse? Jan 22 08:42:45 I can vouch for AS, but gradle builds Android projects just fine without it. Jan 22 08:43:00 ayushkr19, haven't had any problems with AS, and the switch from Eclipse made me very happy Jan 22 08:44:13 so if you'd start new on a long-run project you'd suggest AS+gradle? Jan 22 08:44:48 <_rm> we're currently using as+gradle at work Jan 22 08:45:11 I'm not using gradle, I just imported my Eclipse project into AS Jan 22 08:45:18 fancy42: Yes. With gradle you can use any ide/editor, so you can use AS if you want, but you do not have do. Gradle is doing the actual buliding. Jan 22 08:52:23 ok, but is AS a huge improvement to ADT? Jan 22 08:53:09 I tried it, but went back to eclipse, but I think that AS has a lot of potential, and is already probably better than eclipse Jan 22 08:54:25 ok. but for build management, gradle is your definite suggestion? Jan 22 08:57:13 Also I am a little confused as how to properly setup my testing project structure. All the SDK-specific tests need to run on the device and I need a testing project for this. But do I put the normal Unit Tests in this test project, too? Jan 22 09:33:15 what is difference between pixel and dot ? Jan 22 09:38:46 You mean dp? Jan 22 09:38:55 pixels are the smallest unit of "measure" on a display and a dot could be 1 pixel(depends on the size of the pixel) or multiple pixels Jan 22 09:43:46 anyone here connect via a bouncer like ZNC? Jan 22 09:55:14 how do i read log files with the new monitor app? Jan 22 09:55:23 ddms just crashes and tells me to use monitor.bat Jan 22 09:55:33 (i haven't done android for a long time:P) Jan 22 10:06:09 my app contains a Tab Category (on the ActionBar) which will load the CategoryFragment right under the Tabs. Now i want to click on the detail button in the CategoryFragment to open the DetailFragment right in place of the view of the CategoryFragment Jan 22 10:06:15 how could I do that ? Jan 22 10:15:28 Hi! When hooking a galaxy samsung s4 (android 4.3) to a MacOSX 10.9 via USB, I get the authorization dialog, and I click on ok, but adb devices still list my device as unauthorized. How comes? Jan 22 10:17:24 working around with Transactions at http://developer.android.com/guide/components/fragments.html#Transactions and I can load the new fragment Jan 22 10:17:49 but the whole activity lost its state (no actionbar, pager etc) Jan 22 10:19:52 ogamita disconnect usb cable and connect again Jan 22 10:20:24 if dont work i see 3 option Jan 22 10:20:37 did it, several time. It asks again authorization, I give it again. Jan 22 10:20:59 That said, since tcpip was activated from linux box, I can connect via wifi. Jan 22 10:21:01 restart the macbook check if are some software for the specific phone use depuration by wifi Jan 22 10:21:15 Yes, the later works. Jan 22 10:21:52 What's funny, is that I have the same problem when connecting an iPhone to a linux box, it keeps asking for authorization and doesn't register when I give it :-) Jan 22 10:22:21 I guess they check for same vendor, and break it on purpose when it's the competition. Jan 22 10:22:46 dunno :D Jan 22 10:28:06 is it okay to use asynctask in business loci? Jan 22 10:28:45 *logic Jan 22 10:31:06 isn't the best aproach Jan 22 10:31:27 if use asyntask to handler someting in layout Jan 22 10:31:45 it must independent of logic Jan 22 10:31:53 must be* Jan 22 10:34:35 ok Jan 22 10:34:50 this is to make unit tests independent of the android implementation? Jan 22 10:35:14 unit tests should be in an entirely different project Jan 22 10:35:26 ok Jan 22 10:36:03 but my question was aiming at: if i use asynctask in a business component, i can only test this component on the device, and not locally, right? Jan 22 10:36:36 You can run an emulator, or what would be better is if you use some sort of mocking framework Jan 22 10:36:47 Which fakes the responses. Jan 22 10:37:04 We've got three testing projects in relation to the main project that we publish. Jan 22 10:37:31 thanks. Jan 22 10:38:15 One is just JUnit tests on librarys (not android libraries), they run in seconds. Another is automation tests, which run on an emulator, and take forever, but does mean you get UI Testing done. The third is legacy projects using Robolectric. Jan 22 10:45:08 Please, I'm using a lib, based on another lib. here is my console error : http://pastebin.com/aC3iG1XW Do you have any idea ? I'm getting mad... Jan 22 10:48:11 Might have a problem with two different versions of the Android Support library. Jan 22 10:48:12 Ange_blond, it seems you have added the support library to your classpath Jan 22 10:48:27 twice Jan 22 10:48:44 Desarun: thanks i'll keep that in mind setting up the environment Jan 22 10:49:16 fancy42 I'd follow pduin's lead, looks more likely. It's trying to add the same class twice to the same namespace Jan 22 10:49:51 Desarun> pduin > I did not by myself. I have android-support-v4.jar in a lib project, in "libs", but I don't know how to remove, if I should, etc... Jan 22 10:50:55 Ange_blond, check your project structure, check which modules import that library, and check if it's imported multiple times in all of your project Jan 22 10:51:56 pduin> I never added it, so I asusme it cames by itself with my external lib. So maybe I can remove it from each or my projects ? Jan 22 10:52:37 It's worth a try Jan 22 10:55:01 hum, I'm gonna burn this projects and computer, just to be sure ^ Jan 22 10:55:26 pduin> thanks Jan 22 10:55:43 no problem Jan 22 10:56:02 * pduin gives Ange_blond a jerrycan of gasoline Jan 22 10:56:16 :) Jan 22 12:06:31 Does anyone have any idea of this http://stackoverflow.com/questions/21280794/error-connectionresultstatuscode-internal-error-resolution-null Jan 22 12:44:59 So.... made some cleanup in the code and libpath. Now starting... and crashing! http://pastebin.com/85cQeAU1 Jan 22 12:45:15 I you have any idea... I offer the coffee ;) Jan 22 13:15:45 <[A3G1S]> hey guys, can any 1 help me with keyboard ? I am trying to resize keyboard... Jan 22 13:26:32 [A3G1S] resize android default keyboard? Jan 22 13:26:41 <[A3G1S]> no Jan 22 13:26:51 <[A3G1S]> I am trying to write a keyboard Jan 22 13:27:19 out off my league Jan 22 13:27:37 soz Jan 22 13:29:53 <[A3G1S]> its ok Jan 22 13:30:53 Mornin'! Jan 22 13:31:53 ironmarx: Morning! Jan 22 13:37:18 arunganesan: ironmarx Morning!! Jan 22 13:47:55 <_95A31_> Hi guys, can someone explain to me the difference between branch jellybean and branches cm-1X.X about wpa_supplicant configuration? I see many changes e.i new wi-fi direct interface p2p0, the options -O/data/misc/wifi/sockets and -g@android:wpa_wlan0 etc Jan 22 14:05:25 _95A31_, wrong channel read topic Jan 22 14:06:14 DialogFragment crash if i change orientation before it being launched by Fragment (fragment method called by Asnctask). Do you have any soultion? Jan 22 14:08:03 yoavst: can you pastebin your stack trace? Jan 22 14:08:32 <_95A31_> lemonxah: with "custom ROM" is meaning AOSP roms ? Jan 22 14:08:34 TRomeo: Logcat? Jan 22 14:08:42 yoavst yep Jan 22 14:08:53 Sec, I'll open Eclipse Jan 22 14:09:00 android is possible to drag and drop control when development? Jan 22 14:10:00 _95A31_, any rom this is android app development not rom development Jan 22 14:10:33 fly2web, yes but its not preffered for performance Jan 22 14:11:01 not preferred? only hard coding in XML? Jan 22 14:11:04 lemonxah: Jan 22 14:11:13 <_95A31_> lemonxah: Wops.. My bad >.< Jan 22 14:12:17 <[1]jonc> opinion: sending a facebook auth token for my app (that allows me to view the users interests), should I send over ssl or is that not such a huge deal? Jan 22 14:12:51 TRomeo: http://pastebin.com/LdTY6Fr5 Jan 22 14:13:04 I mean I could always encrypt it myself Jan 22 14:13:15 fly2web, xml isn't hard coding .. that drag and drop feature generates the xml but for performance its better to write the xml yourself instead of having it generated Jan 22 14:13:26 cause of the complexity of some views Jan 22 14:14:46 ok thanks lemonxah, i used to making iOS, there is HIG, but i don't know android well. Jan 22 14:15:10 Hello, Is someone familiar with AndroidQuery lib? Jan 22 14:15:30 I would like to know how to stop image loading/processing? Jan 22 14:15:46 yoavst: you have a null pointer exception in your setContext method> Probably this is happening because android recreates the activity/everything in the activity on a rotation change Jan 22 14:15:56 you'll want to look at the configChanges attribute here Jan 22 14:15:57 Yes, I know it Jan 22 14:15:57 http://developer.android.com/guide/topics/manifest/activity-element.html Jan 22 14:16:11 I read about it and people don't suggest it Jan 22 14:16:14 and possibly the setRetainINstanceState method in your parent fragment Jan 22 14:16:52 it can cause memory leaks if you aren't careful Jan 22 14:16:52 idk what to put in onConfigurationChanged Jan 22 14:16:55 I'll try setRetainINstanceState on parent fragment. tried on the dialogFragment but it didn't help Jan 22 14:17:25 yoavst: I mean, don't just try things and see what sticks. Read about what these things do and figure out what will be best for you Jan 22 14:19:01 I read about put the asynctask inside a fragment, but i don't think that it will work on my task Jan 22 14:19:13 http://www.androiddesignpatterns.com/2013/04/retaining-objects-across-config-changes.html Jan 22 14:19:54 i wrote this a while back http://blog.callumtaylor.net/handlingasynctasksanddialogfragmentswithorientationchangingjava-android Jan 22 14:20:07 may be of help Jan 22 14:20:42 Good morning all. It's been a few years since I've done any android development and it seems I've done something stupid in my manifest file. I can't get my activity named properly to allow it to launch, where I get a concatenation error because the VM is prepending the package name to the already fully qualified activity name http://pastebin.com/NpnucBJZ What's the proper activity naming syntax within a manifest file to avoid this? Jan 22 14:21:14 Can I define other sourceSets than main and instrumentTest in Android Studio? AS refuse to "see" integrationTest as a sourceSet. Jan 22 14:22:53 CallumTaylor: what is WeakReference? Jan 22 14:23:03 a reference that is weak Jan 22 14:23:24 so what does it give to the refernce? Jan 22 14:23:26 http://developer.android.com/reference/java/lang/ref/WeakReference.html Jan 22 14:23:28 Is there a way to sync only one specific device from AOSP instead of the whole branch? Jan 22 14:23:31 yoavst a WeakReference is a reference to an object that allows itself to be garbage collected Jan 22 14:23:40 so the memory won't leak Jan 22 14:23:53 ok Jan 22 14:24:07 i use it so i don't accidentally leak the fragment context Jan 22 14:26:51 CallumTaylor: doInBackground is null on your exmple but shouldn't be null on my code, right? Jan 22 14:27:18 your async task can do what ever, its irrelevant Jan 22 14:27:34 it returns null because the param is Void Jan 22 14:27:44 ok Jan 22 14:30:19 CallumTaylor: I start the AsyncTask from fragment. What should I do then? TaskHelper.getInstance().addTask("task", t, getActivity()); ? Jan 22 14:30:49 i don't know what it is you're trying to do Jan 22 14:31:29 I use tab navigation, so all the code is in the fragment Jan 22 14:32:25 all my example does is make sure when the asynctask finishes, it has a reference to the fragment when the configuration changes Jan 22 14:33:25 but you start the asynctask from an activty Jan 22 14:33:41 and I need to start it from fragment Jan 22 14:36:41 CallumTaylor? Jan 22 14:36:42 right so you adapt it to work with fragment Jan 22 14:37:06 how? I send the user to Image picker, and then the orientation can changed Jan 22 14:37:13 so getActivity() will be null Jan 22 14:37:20 correct Jan 22 14:37:30 you can only deal with the result once its back Jan 22 14:37:44 you could create a wait so it will wait until something gets re-attached before executing a runnable Jan 22 14:38:01 anyway i'll brb Jan 22 14:38:06 ok Jan 22 14:38:51 What about place the fragment inside the activity Java file, and then call methods that will run on the activity? will it work? Jan 22 14:39:32 hm, is my memory wrong or is New->Activity when right-clicking on a package in project explorer gone in latest android studio? Jan 22 14:41:29 yoavst: you don't want to do that, because then your fragment will have a strong reference to your activity, preventing it from being garbage collected Jan 22 14:43:35 TRomeo: can I make a weak refernce between the fragment and the activty and use the Activity context even if orientation changed? Jan 22 14:46:41 yoavst: maybe. The problem s that you get a brand new activity when you change orientations Jan 22 14:46:58 and also a brand new fragment unless you call setRetainInstanceState(true) Jan 22 14:47:28 so f you started the asynctask before the orientation change, and it has a reference to the original activity, it's not going to work Jan 22 14:47:59 yoavst: the point is, when you orientate, the task gets re-attached to the task helper, so get activity won't be null Jan 22 14:48:00 Is there anyone here using kendoui? Jan 22 14:48:16 it'll only be null in-between the last ondestroy and the next oncreate Jan 22 14:48:44 there is a chance it'll finish executing the tasks when that's happening, but it's very small Jan 22 14:49:16 CallumTaylor: So i should force same orientation on downloading? Jan 22 14:49:26 no Jan 22 14:49:49 it doesn't matter whats happening in doInBackground because thats not running on the UI thread Jan 22 14:50:02 the problem comes when you want to update something on the UI in onPostExecute Jan 22 14:50:34 Hi guys/gals. Could you please help me out with the BOOT_COMPLETED behaviour for android >3.1 ? Do I understand correctly, that in order to receive such event, users must run the app once upon installation? Jan 22 14:51:00 yes Jan 22 14:51:11 My onPostExecute call an interface that being called on the fragment. So the launching of the DialogFragment is from the Fragment Jan 22 14:51:34 can you post your code so we can see what it is you're trying to do Jan 22 14:51:47 will it be delivered upon all reboots thereafter? Jan 22 14:52:09 yes Jan 22 14:52:30 CaptainFarrell: Which of the code? the asynctask, fragment (launch of the DialogFragment)? Jan 22 14:52:45 both Jan 22 14:52:58 CaptainFarrell, yep Jan 22 14:53:07 Unless they force stop it Jan 22 14:53:21 thanks. i was fearing that google decided to kill the whole intent-after-reboot functionality by forcing users to run apps after reboot Jan 22 14:53:41 and now I feel like a moron. Thanks everyone for your clarifications! Jan 22 14:55:22 CaptainFarrell: http://pastebin.com/udZjau88 Jan 22 14:56:44 i have to jump into another meeting so i'll look in a bit Jan 22 15:01:40 CallumTaylor|AFK: Ok, Since I'll maybe slove it or disconnect, Hangout me on yoav.goop@gmail.com if i won't be here Jan 22 15:05:02 Hello, how much bonus memory the command "android::largeHeap = true" give to the application ? Jan 22 15:06:35 macTAR: depends on the device, but don't count on much Jan 22 15:07:22 ok Jan 22 15:14:31 So TRomeo: it seem that including the fragment on the Activity file is the best solution Jan 22 15:14:59 you should make the Fragment a static class then Jan 22 15:15:12 if it's an inner calss it will keep a strong reference to the activity Jan 22 15:16:00 why static? Jan 22 15:19:44 CallumTaylor http://pastebin.com/udZjau88 Jan 22 15:20:43 right so because you're using setRetainInstance, you can have a single reference to the fragment in your async task, but you'll have to check to make sure that the fragment is attached to the activity Jan 22 15:20:55 when calling UploadImageFinished Jan 22 15:21:11 and if its not attached, you'll need to wait for it to get attached before executing the method Jan 22 15:23:51 Using of SystemClock.sleep ( or whatever it is) freeze the ui, So how can i do it without freezing? Jan 22 15:25:36 never sleep on the ui thread Jan 22 15:25:40 wait in the asynctask Jan 22 15:26:10 but how, if i need to check if the fragment is attached? Jan 22 15:26:24 to check that i need to be in the ui thread Jan 22 15:27:13 oh Jan 22 15:27:13 uh Jan 22 15:27:19 well I missed the rest of the conversation Jan 22 15:27:29 XD Jan 22 15:27:32 but maybe you can just stash the result somewhere, and the fragment retrieves it when it gets attached? Jan 22 15:27:39 there should be some onFragmentAttached or something, right? Jan 22 15:27:50 but idk if the orientation changed before or not Jan 22 15:28:16 maybe i can check it on the fragment, and if it not attched, put an boolean to true Jan 22 15:28:22 and check it on onAttached Jan 22 15:28:25 what do you think? Jan 22 15:28:37 yeah, that's what I was going to say, put one bit of state somewhere Jan 22 15:28:51 and then test all the corner cases :p Jan 22 15:29:12 Further testing on the activity naming spec issue yields the folowing even odder results: http://pastebin.com/0877jdcv Jan 22 15:29:43 well, i'll try it durka42 Jan 22 15:30:11 but should it be on OnAttach? Jan 22 15:30:37 or maybe later on the lifecycle Jan 22 15:30:48 <— not an expert on fragment lifecycle Jan 22 15:32:19 I think it should be on 4. onActivityCreated(Bundle) tells the fragment that its activity has completed its own Activity.onCreate(). Jan 22 15:35:47 Hello :) I am playing around with Android and iText. Now I want to code this part of my app [http://pastebin.com/mVf9x0v8] so that it saves foo.bar in the internal storage. What do you suggest? Does it work just by replacing "Environment.getExternalStorageDirectory().toString()" with "getFilesDir()"? Also, how can I shorten the code - there are so many convertations, but I don't know how to make it easier :> Thanks Jan 22 15:36:00 durka42: but theoretically, should I open DialogFragment after the Activity OnCreate, or it will be enough to do that after activity attached to an already created fragment? Jan 22 15:43:47 why can't i put environment variables in local.proerties? Jan 22 15:50:29 why would you... Jan 22 15:53:40 you can Jan 22 15:54:33 durka42: I noticed that the onCreate of the Orientation changed happens before the asnctask finished, but getActivity() is still null on the implemented interface on the fragment. Jan 22 15:54:54 what's Orientation, an Activity? Jan 22 15:54:58 what about onResume Jan 22 15:55:55 when the orientation of the device changed, it happens before the asynctask end Jan 22 15:56:09 ugh Jan 22 15:56:15 but getActivity() is null on the implemented interface on the fragment Jan 22 15:56:28 and i can't restart fragmen without activity :( Jan 22 15:57:54 getActivity is null until onAttach Jan 22 15:58:15 do not call container methods in illegal lifecycle states Jan 22 15:58:57 pfn: but if onCreate happend, then onAttach also happend Jan 22 15:59:07 onCreate of what Jan 22 15:59:19 fragment Jan 22 15:59:37 then getActivity is not null Jan 22 15:59:48 but it is null Jan 22 15:59:53 then you're doing something wrong Jan 22 16:00:54 maybe it because if the implemented interface? Jan 22 16:01:04 there is no because Jan 22 16:01:07 your fragment isn't attached Jan 22 16:01:44 I'll try debug onAttach Jan 22 16:03:39 onCreate shouldn't be called on a retained fragment after an orientation change Jan 22 16:04:10 (I assume you're retaining it if you expect to be able to use the same instance after configuration changes Jan 22 16:04:12 ) Jan 22 16:04:16 I removed the setRetainInstance Jan 22 16:05:41 and the onAttach fired before the asynctask finsihed Jan 22 16:05:44 and it null Jan 22 16:05:48 i'll try to save the activity Jan 22 16:06:40 Uhm, if you're not retaining the fragment then it's destroyed along with the activity Jan 22 16:06:51 A new instance will be created Jan 22 16:09:20 I don't really care about that, i want to create a fragmentdialog only XD Jan 22 16:09:24 <_genuser_> geez. how annoyign can google be? Jan 22 16:09:37 <_genuser_> We were unable to verify the account information for your recent purchase with Google Wallet. As a result, your account has been temporarily suspended and your recent order has been cancelled Jan 22 16:09:51 pfn: because in the local.properties is defined where to find the android sdk Jan 22 16:10:04 so it differs from machine to machine Jan 22 16:10:06 <_genuser_> what's to verify? you pay the dev fee and list an app in play store. does google need to know who you are, you child's name, your home address? Jan 22 16:10:12 <_genuser_> anybody seen this before? Jan 22 16:10:30 fancy42, typically, using local.properties isn't very good Jan 22 16:10:36 fancy42, set ANDROID_HOME Jan 22 16:10:42 i did Jan 22 16:10:52 then what's the problem Jan 22 16:11:49 well pfn eventough i set the Activity to a variable on onAttach, it still null on the interface called after the asynctask end Jan 22 16:12:19 and in the debugging it stops on the onAttach before the interface after the asynctask Jan 22 16:12:41 yoavst, you should never set the activity to a variable... Jan 22 16:12:46 yoavst, then it probably got detached Jan 22 16:12:53 before the asynctask Jan 22 16:13:14 well, I'll debug the OnDetach too :( Jan 22 16:13:23 You're keeping a reference to a fragment that is no longer used Jan 22 16:13:42 or you're using the wrong fragment object Jan 22 16:13:45 as simon says Jan 22 16:13:47 You're not retaining it so it's destroyed along with the activity.. When a new activity is created, so is a new fragment Jan 22 16:14:20 You've just verified that onAttach is called on an instance of your fragment.. Not the instance that your asynctask has a reference to Jan 22 16:14:39 oh Jan 22 16:14:51 so i need to setRetainInstance? Jan 22 16:15:22 any reason why ListView.getSelectedItem() would return null when I can clearly see something is selected? Jan 22 16:15:28 pfn: i see. i'm new to ant. i deleted the local.properties, deleted the property from the build xml and it works, because he replaces the missing property with android_home Jan 22 16:16:16 why use ant... Jan 22 16:16:16 SimonVT: it works, thanks :) Jan 22 16:16:31 you have gradle, maven and sbt which all do a much better job Jan 22 16:18:08 BTW - is Adb wifi need drivers? Jan 22 16:18:13 jonc-1, your selectionMode is none Jan 22 16:18:17 yoavst, no Jan 22 16:18:29 so i can run it on every computer? Jan 22 16:22:31 pfn, set to singleChoice, and still null, but thanks Jan 22 16:22:53 btn1.setEnable(isSet); btn2.setEnable(isSet); // btn1 is disabled and btn2 is enabled. how is this even possible ?74 Jan 22 16:23:28 testing on emulator, do not have a device now Jan 22 16:24:28 you're setting it elsewhere Jan 22 16:24:38 nope Jan 22 16:24:54 ah! Jan 22 16:25:11 nope, set in the XML, and only accessed in a fragment which doesn't set it Jan 22 16:25:12 you are right pfm :/ Jan 22 16:26:06 I was setting it also at onServiceConnected Jan 22 16:26:44 I was about to go mad :/ Jan 22 16:26:46 thanks Jan 22 16:27:59 Well, thanks for everbody here for sloving my problem :) Jan 22 16:33:35 jonc-1: selected item isn't for choice mode, you need to use getCheckedItemPosition Jan 22 16:35:05 adb logcat Jan 22 16:35:27 That's twice now in the wrong terminal. Jan 22 16:36:32 Hi all, I'm looking for a iBeacon library for Android. I found this: http://developer.radiusnetworks.com/ibeacon/android/ but want to know if somebody know/has experience with another one? Jan 22 16:36:42 Desarun: Someone did a full git commit once. Not sure why it's funny, but it's funny. Jan 22 16:36:57 Hey guys, following my question up from yesterday, If there is no internet connection, I'd like to send a post request and receive the result in a notification when I'm out of my app and a network connection is available again. So I'm guessing I'd have to register a broadcast receiver to listen to any changes in the network state, when there is, do I need the app to call a service since the app Jan 22 16:36:58 will not be in the foreground? Jan 22 16:37:14 s/and/when Jan 22 16:45:27 when can you know the values (x, y, width , height) of a view inside a Fragment? Jan 22 16:46:26 I would imagine after the view has been inflated Jan 22 16:46:34 after layout Jan 22 16:46:39 you don't know dimensions until layout Jan 22 16:48:09 is there any way to use ellipsize with maxLength Jan 22 16:48:14 On api11+ you can set an OnLayoutChangeListener Jan 22 16:48:28 Pre-11 you need to set a listener on the ViewTreeObserver Jan 22 16:48:43 To get the size of a view Jan 22 16:55:07 Hey guys, following my question up from yesterday, If there is no internet connection, I'd like to send a post request and receive the result in a notification when I'm out of my app and a network connection is available again. So I'm guessing I'd have to register a broadcast receiver to listen to any changes in the network state, when there is, do I need the app to call a service since the app Jan 22 16:55:07 will not be in the foreground? Jan 22 16:56:42 I believe if you make the network call from the broadcast reciever then you don't need a service Jan 22 16:59:29 jonc-1, oh really? Jan 22 17:01:04 not 100% positive, but I believe broadcast receivers can be run when the app is not in the foreground Jan 22 17:02:19 broadcast receivers should take no time to run Jan 22 17:02:25 if they do, you should punt to a service Jan 22 17:02:29 i.e. network calls must be punted Jan 22 17:05:33 pfn, what does punted mean? and so you're saying I cannot make a network call from a broadcast receiver? Jan 22 17:05:39 yes Jan 22 17:05:52 punted == the BR should just start a service to do its work Jan 22 17:06:13 ah I see Jan 22 17:06:27 what happens if I try to make a post request frmo the BR? Jan 22 17:07:04 NetworkOnMainThread Jan 22 17:07:10 I see Jan 22 17:07:18 you should not do /any/ blocking operations on lifecycle callbacks Jan 22 17:07:20 ever Jan 22 17:07:22 never Jan 22 17:07:30 sometimes? Jan 22 17:08:18 pfn, the app would not need to be in the foreground to be notified of any changes in the network connection right? But will need to be running in the background? I'm guessing I'm going to have to use a binding service? Jan 22 17:08:43 why would you need to bind a service Jan 22 17:08:55 avoid binding services unless you have to do ipc Jan 22 17:09:58 pfn, I see so I would need an IntentService which stops itself after everything is handled? Jan 22 17:12:16 pfn, I will need a result from the network call Jan 22 17:12:34 so I thought I would need a boundservice Jan 22 17:13:26 it's in the background, what do you plan on doing with the result... Jan 22 17:13:30 pop up your ui? Jan 22 17:13:38 and you don't need a bound service to communicate results Jan 22 17:13:49 and if it were bound, what is your intention, block until you receive a result? Jan 22 17:13:58 all paths lead to darkness Jan 22 17:14:46 pfn, with the result I need to use to open up an app Jan 22 17:14:57 so a normal IntentService would be fine in this instant Jan 22 17:15:03 instance Jan 22 17:15:12 so, you plan on annoying the user by popping your app up out of nowhere... Jan 22 17:15:44 pfn, haha no I mean when there is a connection it would send a notification and when that notification is clicked open up an app Jan 22 17:15:57 ok Jan 22 17:16:17 but the notifcation would contain the result Jan 22 17:17:06 Surely there is a way to ellipsize a TextView based on its maxLength... (see what I did there) Jan 22 17:18:58 pfn, so an IntentService is fine in this case and not a boundservice? Jan 22 17:19:02 is there anyway to stop android studio adding "android" in front of R. when R hasn't been generated yet? Jan 22 17:19:30 they fixed that in 0.3 Jan 22 17:19:39 file a bug if it still happens Jan 22 17:19:45 hm I'm on 0.3.7 Jan 22 17:19:51 ok maybe it was 0.4 then Jan 22 17:19:57 drat Jan 22 17:20:11 ok i'll give it a go, it was only a matter of time until i had to update anyway Jan 22 17:20:42 [summary from my question a week or so ago about OAuth and google play services... http://sbktech.blogspot.com/2014/01/inside-android-play-services-magic.html - if you spot goofs, please do let me know :-)] Jan 22 17:20:52 if R hasn't been generated yet, why wouldn't it add android. in front... Jan 22 17:20:55 sounds smart Jan 22 17:21:00 no its not Jan 22 17:21:01 trust me Jan 22 17:21:34 yeah it's not smart Jan 22 17:21:44 it's annoying because it's never what you want Jan 22 17:21:57 but you don't have R... Jan 22 17:22:16 Why onSaveInstanceState didn't call in my fragment? Jan 22 17:22:36 you don't have R because your project has errors, your project has errors because it keeps putting "android." in front of your R Jan 22 17:23:02 ok, then that's because android studio is dumb Jan 22 17:23:49 hence why it was fixed Jan 22 17:24:46 pfn, I'm guessing IntentService? :S Jan 22 17:27:57 I need more ram Jan 22 17:29:07 downloadmoreram.com Jan 22 17:30:10 gr issue still exists in 0.4.2 ;_; Jan 22 17:32:54 CallumTaylor: file a bug. I'm not seeing it happen, though. Jan 22 17:33:05 yeah will do Jan 22 17:35:39 Anyone have a rough idea where junit(non-instrumentation) test support is at in order of importance for android studio? Thats actually one of my main reasons I'm not going to it full time. Jan 22 17:36:55 The gradle plugins and all are great but I'd love AS to support it Jan 22 17:39:50 I have a public class that extends AsyncTask for sending web requests but I want to have a callback inside the calling class Jan 22 17:40:12 Hi, I'm getting crazy with an oddity about SeekBar using custom drawable for progress and thumb. Like you can see in linked snapshot when I put SeekBar.position value to 0 the thumb is drawn out of progress bar. I wasn't able to understand what parameter set to align correctly thumb size with progress place... have you any suggestion ? https://www.dropbox.com/s/e3fsx4vbqggcevh/Screenshot%20-%2001222014%20-%2006%3A37%3A06%20PM.png Jan 22 17:41:39 Any recommendations for how to do it? Am pastebinning my current work now Jan 22 17:42:27 I put an app on the marketplace with a certificate to last 24 years. What should I do If I need it to last longer. I can't change the certificate now without having to create a whole new app package name… right? Jan 22 17:42:32 conor_f, usually I use Handler messages to do that Jan 22 17:43:05 http://pastebin.com/5PtQ1sdP Jan 22 17:43:16 Handler messages? Jan 22 17:44:10 conor_f, perhaps I've mistake your question (I'm reading your code to better understand) Jan 22 17:45:45 At the moment I've a base class (APICallingActivity). I store that in the constructor, then try to access it onPostExecute Jan 22 17:46:35 ah ok ... why you don't pass a Handler reference and just post a message to the caller ? Jan 22 17:47:11 Because this is my first time hearing about Handlers Jan 22 17:47:30 Googling now Jan 22 17:47:31 ah ... Jan 22 17:47:48 ASyncTask is a mix of Thread + Handler Message Jan 22 17:48:51 Pratically if you have a Handler = new Handler() { .... and you implement a getMessage() } you can send this Handler to every one (also a thread) and get the message when you job is done... Jan 22 17:49:03 git config alias. Jan 22 17:49:17 Oh ding dang it. I do that, too. Pardon the mav, folks. Jan 22 17:49:35 in 24 years you wont be using the same certificate Jan 22 17:49:57 If I want to create a wrapper class that uses bindService should I extend it using something or pass the context from the class that extends it? Jan 22 17:50:02 Greetings, folks! Can anyone tell me what's wrong with samsungs? I had release-like app and samsungs behave so weird that I can't even understand sometimes what appears on the screen -.- For instance, I have a ListView with kinda light blue background color which is filled with rows as relative layout with white background color, but no! Every row is same backgrounded as it's list view. Is it ok? Jan 22 17:50:27 shineworld: hmm. Handler messages seem to be used for progress bars, etc? Jan 22 17:50:31 the real question is, what's right with samsungs Jan 22 17:50:46 conor_f, in this stupid sample (https://github.com/shineworld/uartaccessory/blob/master/android/UartAccessoryTest/src/it/shineworld/uartaccessorytest/MainActivity.java) you can see how anyone (also a thread) can send data to UI thread using Handler Message Jan 22 17:50:51 you can use it to send any type of data Jan 22 17:51:00 CallumTaylor: okay, can you point me then what did I wrong? Jan 22 17:51:05 Check public void handleMessage(Message msg) Jan 22 17:51:55 no idea, samsungs are annoying Jan 22 17:52:03 could be a list caching issue Jan 22 17:52:07 Another way is to create an interface IAsyncTaskCaller with one methond public method() and implement it in your Activity then send it to your custom ASyncTask Jan 22 17:52:48 which is what you have done Jan 22 17:52:50 shineworld: that's sort of what I'm trying to do with the base class Jan 22 17:52:51 yeah Jan 22 17:53:03 then why isn't it working if that's a valid way of doing it? Jan 22 17:53:20 instead of use a class use an Interface Jan 22 17:53:32 should work Jan 22 17:53:49 another cool thing that there is some linear layout with dark gray background color with two linear layouts inside, left one is same backgrounded, second one is white, but on samsung right one is almost the same as its parent :\ Jan 22 17:54:10 absolutely the same* Jan 22 17:54:38 why would it work with an interface and not a class? Jan 22 17:55:06 mikedg: you think? Jan 22 17:55:36 Why in fragments not called onSaveInstanceState? Jan 22 17:55:58 sure - all the prime-number/discrete-log based certs would be made obselete by quantum computers :-) Jan 22 17:56:10 *obsolete Jan 22 18:17:14 Should I start SettingsActivity with ActivityFromResult() and then apply the settings on return, or should I start it with StartActivity() and then immediately apply the settings when changed? Jan 22 18:35:00 ok so im creating a minimap on the fly at run time where one pixel in the mini map equates to one tile. as the player moves around the map the minimap is filled in. well i save this out as a PNG and i notice that its dithered. BADLY Jan 22 18:35:33 i know its not the compression causing this because png is lossless so it must be happening as i write the pixels into the minimap bitmap Jan 22 18:36:36 Is it uncommon to have basically just one activity and do everything else with Fragments? I need a fragment at the bottom of all screens which persists data across screens. Jan 22 18:42:40 in Bitmap.setPixel(int x, int y, int color) the color must be a non-premultiplied ARGB value. <-- what is pre-multiplied? Jan 22 18:43:07 no idea Jan 22 18:43:23 all I know Color is aarrggbb Jan 22 18:43:42 bricker`LA, no Idea Jan 22 18:44:03 why is the bitmap im creating one pixel at a time (sort of) dithered? Jan 22 18:45:19 miniMap.setPixel(mx, my, pel); Jan 22 18:45:35 oh. is it because the bitmap ix 656 not 8888 ? Jan 22 18:46:03 let me test that! Jan 22 18:48:10 nope. weird. ws RGB_565, now ARGB_8888 and the bitmap is still being dithered as i construct it Jan 22 18:51:56 I440r: is bitmap bigger or smaller than area its painted on>? Jan 22 18:53:24 the main map is 512 by 256 tiles. the bitmap for the minimap is 512x256 pixels Jan 22 18:53:48 i only display a 48x48 window on it which is scaled but the minimap bitmap itself is not scaled Jan 22 18:54:00 no idea then Jan 22 18:54:02 its being dithered as i write pixels into it one pixel at a time Jan 22 19:00:28 just idly curious really - but posting code/bitmap output might offer some clues (or at least, satisfy my curiosity :-) Jan 22 19:03:56 btw, pre-multiplied just means the alpha value is already multipled into the rgb values. [eg: 255,255,255 with alpha 50% would be 127,127,127,0.5]. Don't do that. Jan 22 19:04:40 so dont put the alpha in there? Jan 22 19:05:02 do i specify the alpha as zero then? Jan 22 19:05:09 when i do that i get no pixels lol Jan 22 19:05:46 erm no im not ybderstabdubg te 127,127,127,0.5 Jan 22 19:05:51 understanding Jan 22 19:05:57 blargh i need to L2 Type lol Jan 22 19:06:17 im specifying my colors as 0xffRRGGBB Jan 22 19:06:18 just do everything as you'd do normally when specifying colors as an argb_8888 int. Jan 22 19:06:23 that's fine Jan 22 19:06:43 well im drawing pels into the bitmap and saving the bitmap as a PNG. the resultant file is dithered Jan 22 19:07:01 see above to get more useful help :-) Jan 22 19:07:08 its not supposed to be. its not a scaled image, its 512x256 ARGB_8888 Jan 22 19:07:15 premultiplied just means the colors has already been multiplied by the alpha Jan 22 19:07:35 this code i can post, plus the bitmap. gimme a min Jan 22 19:11:52 http://bpaste.net/show/171002/ <--- source Jan 22 19:12:07 http://bpaste.net/show/171004/ Jan 22 19:12:20 png showing discovered area dithered Jan 22 19:12:43 ugh stupid bpaste lol Jan 22 19:12:57 where can i paste the PNG file so that it displays as an IMAGE lol Jan 22 19:13:05 Guys, do you know the best SMS Gateway to use in my android app ? Jan 22 19:13:47 I440r: imgur? Jan 22 19:14:10 I440r: snag.gy Jan 22 19:16:03 ok snag.gy is totally USELESS Jan 22 19:16:17 When i'm trying to send image to imageshack api using apache httpclient (from external jars) it fail with this error - java.net.SocketException: sendto failed: ECONNRESET (Connection reset by peer). It happens only where the image is from whatsapp folder. do you have any idea why? Jan 22 19:16:18 it pastes it as a JPEG which is lossy and thus destroys the dithereing Jan 22 19:17:13 the folder is - /storage/emulated/0/WhatsApp/Media/WhatsApp Images/IMG-20140122-WA0004.jpg Jan 22 19:17:51 and imgur does not seem to be a PASTE site for images, its a share site for images Jan 22 19:18:25 well.. you normally don't *paste* images... Jan 22 19:18:37 well i got an idea :P Jan 22 19:18:38 not into a web browser at least Jan 22 19:19:09 someone? Jan 22 19:19:30 ironrock: there's plenty. best depends upon your needs. Jan 22 19:19:32 www.isforth.com/minimap.png Jan 22 19:19:45 dont view it in the web browser, the browser seems to smoothe the image out Jan 22 19:19:54 ping JakeWharton - got a Spoon q for ya before i post up q's on github commits Jan 22 19:19:57 save the PNG and look at it in the gimp Jan 22 19:20:07 tophyr: go Jan 22 19:20:21 kbs see my code? Jan 22 19:20:27 StackOverflow is for questions, not GitHub Jan 22 19:20:59 SpoonRunner.Builder has an option to set the classpath - i'm wondering what it might be needed for. i'm looking at spoon-gradle-plugin and it sets it, i *think* (don't know groovy/gradle well enough to say certainly) it sets that classpath to its own self, which seems pointless Jan 22 19:20:59 I440r: yes - and the first thing I wonder is whether the calculation being done is causing the effect. dunno Jan 22 19:21:19 the calculations on the pixel colors? Jan 22 19:21:26 i tinkered with that and removed the line setting the CP in the Builder and the tests still ran fine. Jan 22 19:21:42 ok well all i am doing is adding the RGB of the ground layer tile to HALF the RGB of the object layer of the tile Jan 22 19:22:15 so (r1 + r2/2) + (g1 + g2/2) + (b1 + b2/2) Jan 22 19:22:30 the masking is just to make sure the additions of one color to not bleed into the other colors Jan 22 19:22:30 ColinTheShots: Which would you recommend ? Jan 22 19:22:57 ColinTheShots: I've used Twilio, but they're not always the cheapest and they don't yet support all countries Jan 22 19:23:28 oooh erm seems im creating a 565 bitmap in here. let me change that to argb 8888 Jan 22 19:23:30 tophyr: the details escape me, but i know the Maven plugin uses that Jan 22 19:23:32 When i'm trying to send image to imageshack api using apache httpclient (from external jars) it fail with this error - java.net.SocketException: sendto failed: ECONNRESET (Connection reset by peer). It happens only where the image is from whatsapp folder. do you have any idea why? the folder is - /storage/emulated/0/WhatsApp/Media/WhatsApp Images/IMG-20140122-WA0004.jpg Jan 22 19:23:36 ironrock: I'd just investigate the options and compare it to my needss Jan 22 19:24:03 JakeWharton: the maven plugin for gradle uses it? Jan 22 19:24:11 no the Maven plugin for Maven uses it Jan 22 19:24:20 it's used when the child processes are spawned Jan 22 19:24:41 ahh ok. i'll read more of the source and see if i can follow it any better Jan 22 19:24:54 and try to ask more specific q's then :) Jan 22 19:24:54 when invoked in a Maven mojo, we want the new process to have the spoon jar (and its dependencies) as the classpath Jan 22 19:25:15 the current classpath is only maven stuff so it would fail otherwise Jan 22 19:25:37 ohhhhhh.. actually that does make a bit more sense now then Jan 22 19:25:46 the CP is for the *spoon* binaries? Jan 22 19:25:56 i was assuming it needed the CP of the application under test Jan 22 19:27:02 right. it's the classpath for the runner to re-spawn itself Jan 22 19:27:19 gotcha. ok, that makes a lot more sense. Jan 22 19:27:29 if you just run it normally with java -jar the normal classpath is fine. it's only when running inside Maven that we have to "fix" it Jan 22 19:27:54 JakeWharton: is there a way to disable the animation on tabs navigation where you swipe left from the first tab, and swipe right from the last tab on SherlockActionBar? Jan 22 19:28:11 ok. it seems fine in gradle so far but i haven't done anything fancier than running 1 suite on 1 device yet Jan 22 19:28:53 yoavst: no Jan 22 19:29:03 really? it annoying Jan 22 19:29:35 i don't know what you're referring to, but I know there's no options for anything like that in the native action bar (therefore none in ABS) Jan 22 19:29:54 well, in the play store it looks much better Jan 22 19:30:06 that's because they aren't using the action bar tabs Jan 22 19:30:26 so what they use? Jan 22 19:30:39 custom control Jan 22 19:30:42 http://www.pushing-pixels.org/2013/05/09/android-bits-and-pieces-sliding-tabs-madness.html Jan 22 19:31:33 I440r: why is snag.gy totally useless? I use it and it has a purpose for me. Jan 22 19:32:05 JakeWharton: Where do I download CustomControl? Jan 22 19:32:19 I440r: btw - in addition to the config issue you noticed, would you have issues if r1+r1/2 exceeds 255? [eg: 255 on the top + 2 on the ground will; if I read correctly, turn into 0. The result will be strange colors, I guess. Not sure exactly what you meant by dithering...] Jan 22 19:32:24 SimonVT: https://developer.apple.com/ Jan 22 19:32:28 ty Jan 22 19:32:52 doge_wow, not toally useless, but uses JPEG when you paste into it which smooths out the dithering im trying to fix Jan 22 19:33:10 kbs its masked to 0xff 0xff00 0xff0000 for each color Jan 22 19:33:13 bgr in that order Jan 22 19:33:19 JakeWharton: i need a library that applies high jpeg compression to all data Jan 22 19:33:23 so thers no bleeding from one color into the next Jan 22 19:33:27 I440r: yes, but that wasn't exactly what I was saying :-) Jan 22 19:33:30 brb, making toast Jan 22 19:33:35 erm ok let me read again lol Jan 22 19:33:43 mikedg: http://msdn.microsoft.com/en-us/default.aspx Jan 22 19:34:07 When i'm trying to send image to imageshack api using apache httpclient (from external jars) it fail with this error - java.net.SocketException: sendto failed: ECONNRESET (Connection reset by peer). It happens only where the image is from whatsapp folder. do you have any idea why? the folder is - /storage/emulated/0/WhatsApp/Media/WhatsApp Images/IMG-20140122-WA0004.jpg Jan 22 19:34:12 I440r: I think your masking logic causes colors that "overflow" to turn dark, rather than being truncated to 0xff Jan 22 19:34:15 download the PNG file from www.isforth.com/minimap.png and look at it in the gimp Jan 22 19:34:31 kbs hang on toast is buring lol Jan 22 19:35:11 JakeWharton: I'm new here. Are you the man of ActionBarSherlock ? Jan 22 19:35:40 ironrock: he is gender neutral Jan 22 19:35:45 :) Jan 22 19:36:33 lol, his github has a man photo. Jan 22 19:37:27 whatever, i have to say, congratulation for your work ! Jan 22 19:37:54 Environment.getExternalStoragePublicDirectory(Environment.DIRECTORY_PICTURES); return /storage/emulated/0/Pictures Jan 22 19:38:03 it's a pleasure is chatting with you. Jan 22 19:38:10 but the pictuers are in DCIM Jan 22 19:41:15 is there any way to set the style for a bulk of buttons like what we usually did in web 's css (define class and set style for class) Jan 22 19:41:31 it kinda tedious to just copy the name of the style on every buttons I make Jan 22 19:42:05 bubuzzz: http://developer.android.com/guide/topics/ui/themes.html Jan 22 19:42:08 you want themes Jan 22 19:43:57 ironrock: thanks Jan 22 19:44:11 yes i know. But in the end I have to add the style like every time I create a new button Jan 22 19:44:31 just override the android button style Jan 22 19:44:32 bubuzzz: the theme has a buttonStyle attribute for defaults Jan 22 19:44:50 you can also use a ContextThemeWrapper to inflate layouts in bulk using a theme different from the one set on the activity Jan 22 19:45:20 ok changed the 565 to 8888, lets see if that fixes it Jan 22 19:45:32 which is quite convenient for variably themed apps Jan 22 19:46:19 @JakeWharton the ContextThemeWrapper looks promising. Thanks Jan 22 19:48:18 nope its still dithered Jan 22 19:48:40 if you view the png in the gimp and zoom in on the revealed area of the map it has a checkerboard look to it Jan 22 19:49:46 ya sure its not partially transparent Jan 22 19:49:59 because gimp does have checkboard look under png Jan 22 19:51:28 I440r: yes - (and as I mentioned in addition to that overflow bug) not clear to me [at least, immediately :-)] that the color selection in your tileColors won't lead to that. At the very least, don't see anything to indicate some weird issue with the PNG encoder, though that probably doesn't solve your immediate isuse. Jan 22 19:52:45 kbs the water has no layer2 objects over it so its (R + 0/2) + (G + 0/2) + (B + 0/2) Jan 22 19:53:08 and yet the water is still checkerboarded - so overflow is not the problem but yes good catch i should do max not mask Jan 22 19:53:16 i.e. if color is > 255, color = 255 Jan 22 19:53:49 Is this the best channel to ask about GCM problems? Jan 22 19:54:27 From the 3rd part application server end of it. Jan 22 19:55:01 Actually, the problem seems to show up on the Android app as well. Jan 22 19:56:00 Basically, I'm getting 401s, App developer is getting INVALID_SENDER, and we're not sure why. I think I accounted for the possible causes listed in the troubleshooting section. Jan 22 19:56:45 what headers does the server require? Jan 22 19:57:04 Authorization, and perhaps Content-Type. I'm supplying both. Jan 22 19:57:09 I doublechecked the API key. The syntax *maybe* is wrong but I'm using a Python library to make the requests. Jan 22 19:57:34 same thing happens from multiple phones? Jan 22 19:57:36 The docs I think erroneously say I may have sent the wrong Project Number. I tried the project number just in case and it didn't work. (I put in an error report for that) Jan 22 19:58:12 Oh, well the phone part of it is another developer. But I think she's only on one phone, maybe an emulator. But I'll suggest she tries multiple. Jan 22 19:58:22 I440r: *nod* dunno enough about the tile-colors and your index selection to be sure; but I'd be surprised if it isn't pretty generating that checkerboard pattern when you're setting the x/y values. You might want to simply log it for your 32x24 top-left section to confirm it. Jan 22 19:58:23 Me, I'm on the 3rd party server thing. Jan 22 19:58:33 s/pretty/pretty much Jan 22 19:58:57 curl --header "Authorization: $api_key" --header Content-Type:"application/json" https://android.googleapis.com/gcm/send -d "{\"registration_ids\":[\"ABC\"]}" Jan 22 19:59:10 how to I create a 9 patch with choppy triangular shaped edges? Jan 22 19:59:18 As described in the doc. And the error response content is in HTML, interestingly enough, not json. Jan 22 20:00:31 GCM looks to be enabled for our app. Jan 22 20:00:41 is it possible? Jan 22 20:01:00 We haven't set a whitelist, and StackOverflow comments imply that by default everything is allowed. Jan 22 20:01:27 is anybody else bothered that LatLng is not LatLon? I dunno why, it just bugs me. Jan 22 20:01:35 Is it possible to write an app for Android in Python? Jan 22 20:01:35 CocoStorm: unfortunately, you can't create "evenly-spaced" frieze patterns - at best, you could have the same number of triangles per edge, but their angles would be squished as needed. Jan 22 20:01:44 CocoStorm, do you want the corners to stretch? Jan 22 20:02:08 I'm mentally imagining that CocoStorm wants to make a postage-stamp serrated edge Jan 22 20:02:25 kbs, yes pretty much and it's not possible then? Jan 22 20:02:28 Cocostorm, you mean the entire edge? like kbs said: a stamp? Jan 22 20:02:32 nope Jan 22 20:02:33 ya Jan 22 20:02:40 I see Jan 22 20:02:57 so I'd have to make separate png files for each screen resolution? Jan 22 20:03:05 pretty much Jan 22 20:03:10 I see Jan 22 20:03:14 dang Jan 22 20:03:16 you could set the content area of it, but that's about it Jan 22 20:03:33 Cocostorm, there is a pretty simple guideline for creating images for the different DPI's Jan 22 20:03:49 jonc-1, oh yeah? Jan 22 20:03:53 I'll check em out Jan 22 20:03:57 http://stackoverflow.com/questions/13763988/android-device-screen-size-supporting-large-screens Jan 22 20:04:14 but it's probably more than just dpi's is it? I was somehow imagining that this would form the background for arbitrary-sized ui elements Jan 22 20:04:59 Ok the Android dev just got something. Still no good for me. Jan 22 20:05:04 kbs, yep Jan 22 20:05:04 if you make XHdpi first, then Hdpi = 75% of XHdpi, Mdi = 50% of XHdpi, Ldpi = 37.5% of XHdpi Jan 22 20:05:42 yeah that won't easily happen CocoStorm Jan 22 20:06:11 Most of the time you can just make xhdpi, and add higher res and lower res assets as needed Jan 22 20:06:26 I see, heh well I'll have a look at the SO post Jan 22 20:06:40 now you could create all of these images in code ;) Jan 22 20:06:49 as long as it was a solid color / had a border only Jan 22 20:07:01 CocoStorm: hm. If you need arbitrary sized edges, might need to do something else that just dpi-based stuff. I've made a custom (programmatic) drawable once; but that was for a fairly simple shape. If your triangles are easy enough to draw at run-time, maybe that's an option. Jan 22 20:07:13 jonc any idea why a bitmap im creating one pixel at a time is ending up dithered? Jan 22 20:07:27 i posted the code earlier but will do so again as there have been minor changes Jan 22 20:07:43 * kbs continues to suggest that it's because the pixel is being drawn dithered :-) Jan 22 20:07:55 kbs WHY is it being drawn dithered? Jan 22 20:07:57 that would be my first idea :P Jan 22 20:08:02 post the code really quick Jan 22 20:08:08 posting brb Jan 22 20:08:22 http://bpaste.net/show/171025/ Jan 22 20:08:23 don't create ldpi assets Jan 22 20:08:37 I440r: something to do with your index selection/color tiles - but hey - it's easy enough to confirm the theory by logging; and you can debug backwards from there. Jan 22 20:08:41 thanks kbs I'll give it a shot :) Jan 22 20:08:42 hell, we don't even make mdpi assets anymore Jan 22 20:09:13 kbs log what? lol Jan 22 20:09:21 the colors of the pels as i write them? Jan 22 20:09:48 I440r: right. "You might want to simply log it for your 32x24 top-left section to confirm it." Jan 22 20:10:22 if i write a pal for water 0xffRRGGBB theres no layer 2 so its just the RGB of the water tile. when i write ONE water tile pixel the color written should not have any knowledge of the colors of the pixels near it. Jan 22 20:10:45 yea i can do a memory dump of the top left of the map Jan 22 20:11:52 if(tx < 0), change the mapx and the pixelx by 16 Jan 22 20:11:59 unless the png save operation adds artifacts to the saved file (which would be utterly stupid) Jan 22 20:12:13 jonc-1 the MAP coordinates are in 16x16 tile blocks Jan 22 20:12:18 if tx < 0, that means that you want to go back a row correct? Jan 22 20:12:27 go up a row* Jan 22 20:12:43 no it means the x coordinate within the 16x16 block is in a different block Jan 22 20:12:50 or is it just supposed to wrap around to the otherside? Jan 22 20:12:53 gotcha Jan 22 20:12:56 :) Jan 22 20:13:01 so yes. it wraps :)) Jan 22 20:13:20 good q tho, shows your actually paying attention to the code lol Jan 22 20:13:40 do you have a screenshot of the dithering? Jan 22 20:13:57 Should I start SettingsActivity with ActivityFromResult() and then apply the settings on return, or should I start it with StartActivity() and then immediately apply the settings when changed? Jan 22 20:14:02 www.isforth.com/minimap.png dont view in browser, download it and view in gimp Jan 22 20:15:09 so the issue is how you have this like every other block is the wrong color? Jan 22 20:15:43 it has a checkerbord look to it that im trying to understand Jan 22 20:15:48 yeah Jan 22 20:15:55 subtle shading differences Jan 22 20:16:08 its a neat looking affect but its not a desired one lol Jan 22 20:17:21 could it be an alpha problem? Jan 22 20:17:25 I think it is Jan 22 20:17:31 all alphas is 0xff no? Jan 22 20:17:42 let me FORCE that Jan 22 20:19:20 added pel |= 0xff000000; and still checkerboarded Jan 22 20:19:34 could this LOSSELSS png save be buggering up the saved file? Jan 22 20:19:42 heh Jan 22 20:19:58 so much easier to blame the compressor :-) Jan 22 20:20:05 it is 1 of 2 problems it seems, that the alpha is getting changed (which makes the black background come through), or you are somehow changing the value of all every other pixel in a row to be darker by about the same amount Jan 22 20:20:09 1 second Jan 22 20:20:15 kbs i dont honestly believe its that lol Jan 22 20:20:24 google wouldnt get something that fundamental buggered Jan 22 20:20:56 its not getting darker, its alternating light dark sort of. Jan 22 20:21:32 yeah Jan 22 20:24:07 tried saving as jpeg with quality at 10? Jan 22 20:24:16 shouldn't see too many artifacts from that Jan 22 20:24:29 * kbs continues to point out that this is not a compression artifact Jan 22 20:24:50 I don't think it is either Jan 22 20:24:51 jonc i dont want any lossy compression Jan 22 20:25:26 could it be in your tileColors array? Jan 22 20:26:06 the tile colors array is derived from the tile pngs. each tile is 72x72. what i do is for each tile i add up the RGB of every pixel and avg them. thats the RGB of the minimap pixel Jan 22 20:26:21 water is tile 0xNN and nn is the index into the color table Jan 22 20:27:59 of course i precalculate these values :) Jan 22 20:28:32 yes but the average color of even 2 grass tiles could be different (even if next too each other) Jan 22 20:28:46 next to* Jan 22 20:29:16 if you have different shades of grass that is Jan 22 20:29:50 ok i have a PNG that contains all the images for every tile. each tile is 72x72. for each tile i add up all the red, all the gree, all the blue. i then divide each by 72*72 and thsts the RGB value saved for that tile Jan 22 20:29:59 not "that tile in that position in the map" Jan 22 20:31:15 i dont look at the map, just the PNG for the tiles Jan 22 20:31:55 http://www.color-hex.com/color/00178a, http://www.color-hex.com/color/00188a Right next to each other in your tileColors array Jan 22 20:32:04 and those are the 2 shades of blue in the image you showed Jan 22 20:32:35 jonc shallow water, deep water Jan 22 20:32:37 try putting both of those to the same blue and then save it out and check it Jan 22 20:32:40 I understand that Jan 22 20:32:55 u mean make both waters the same color? Jan 22 20:32:58 to test? Jan 22 20:33:28 actually wtf its not happening any more lol Jan 22 20:33:33 :P Jan 22 20:33:49 that always happens Jan 22 20:33:52 all i did was remove the 0xffRRGGBB <-- removed the 0xff alpha part from the color table Jan 22 20:33:59 srsly wtf over lol Jan 22 20:36:05 nope. now its happening again. omfg lol Jan 22 20:36:30 for ONE try it didnt happen. lets back out one change at a time Jan 22 20:38:28 what is layer1 and layer2 for tile Jan 22 20:39:07 c1 += ((c2 & 0xfe) >> 1); Jan 22 20:39:12 is going to remove some alpha Jan 22 20:39:37 hey guys...question Jan 22 20:39:43 thats doing only the B part of the color Jan 22 20:39:47 when I start an intent and create a new activity Jan 22 20:39:52 i take the B for layer1 and add half the B for layer 2 Jan 22 20:39:59 i do the same for read and green Jan 22 20:40:01 is that call to startIntent blocking? Does the other activity get created before it returns? Jan 22 20:40:07 prgmmax: no Jan 22 20:40:09 k Jan 22 20:40:22 kind of didn't think so but wanted to double check because that would make things easier for me right now Jan 22 20:40:26 figured it was worth an ask Jan 22 20:40:27 thanks Jan 22 20:41:35 ok. let me push a new minimap for you to see Jan 22 20:42:13 download isforth.com/minimap.png again Jan 22 20:42:29 you see how its all alpha = 0x00 except where i have discovered... Jan 22 20:42:35 well. thats the fix but not Jan 22 20:43:19 in the miniView constructor where i erase the minimap bitmap i use a paint where i have set the alpha to 0xff and filled with color = black Jan 22 20:43:48 well with this i get the artifact. if i change the paints alpha to zero i dont get the artifact Jan 22 20:44:49 then somewhere the bit manipulating is wrong and dirtying a bit it isn't supposed to Jan 22 20:45:11 since it's ok when 0 but no 1, I would imagine an | is where it is getting bad Jan 22 20:45:46 which | Jan 22 20:46:13 if java cant do | logic then im worried for it lol Jan 22 20:47:28 ppl scoff at my use of (n >> 1) instead of (n / 2) lol Jan 22 20:48:03 <|0xD34D|> do the artifacts show up when you have a map consisting of all the same tile? Jan 22 20:49:05 I440r, it will be optimized down to n >> 1 most likely :P Jan 22 20:49:52 jonc-1, granted. but i for example would never hire a "embedded apps coder" who learned coding in java and migrated to the real world Jan 22 20:50:04 java teaches you BAD MOJO Jan 22 20:50:14 to be fair Java was created to be an embedded language Jan 22 20:50:23 it IS an embedable language Jan 22 20:50:43 its very efficient as an embedded language. whats not efficient is a C/ASM coder who learned coding by learning java Jan 22 20:50:51 haha Jan 22 20:51:04 I haven't done ASM in a long time Jan 22 20:51:09 myvar = foo / 2; <--- oopts that actually did a DIVISION not a shift Jan 22 20:51:22 and C it has been 3-4 years (doing Cell processor stuff on a ps3) Jan 22 20:51:26 i just wrote a NDK library for this game entirely in asm :P Jan 22 20:51:54 well i wrote two ndk libs, one in asm, the other in c Jan 22 20:51:54 I just did my first port of a game on NDK using SDL and OpenGL Jan 22 20:52:06 Although I had to translate OpenGL to OpenGL ES Jan 22 20:52:14 Ologn, not many ppl in here know what the NDK is :P Jan 22 20:52:32 they make a cross sign if you say "ndk" Jan 22 20:52:36 haha Jan 22 20:52:40 https://play.google.com/store/apps/details?id=com.panaceasupplies.android.games.plumber Jan 22 20:52:58 i need to learn opengl Jan 22 20:53:48 arm assembly is risc isn't it? Jan 22 20:53:56 There's OpenGL 1, OpenGL 2, OpenGL 3, OpenGL 4, OpenGLES 1, OpenGLES 2 etc. Then there's C++, Java etc. Jan 22 20:54:38 they say it is risc. i say it USED to be risc Jan 22 20:54:44 its now a bloated HUGE instruction set Jan 22 20:54:48 haha Jan 22 20:54:58 specially if you have anything to do with thumb/thum2 Jan 22 20:55:04 d'oh Jan 22 20:55:05 i will never use thumb2 Jan 22 20:55:26 if ARM mode is deprecated and removed from the core ill stop developing for arm Jan 22 20:55:38 the entire reason for thumb in the first place was theoretically to be more space efficient Jan 22 20:55:46 forth is about 10 times more space efficient than thumb Jan 22 20:56:34 the only VALID reason i can see for thmub1 is on targets such as the nintendo DS where the address buss to the flash is 16 bits. if you coded your games in arm every opcode fetch would require two memory reads Jan 22 20:56:54 thumb2 is a product of the mind of satan and his minions Jan 22 20:58:46 p.setAlpha(n); p.setColor(Color.BLACK); vs.drawRect(r1, p); <-- if n1 is 0xff i get the artifact. if its 0x00 i dont Jan 22 20:59:16 do you separate your Async Tasks into their owns Class files? Jan 22 20:59:25 p is what? Jan 22 20:59:29 a paint Jan 22 20:59:34 gotcha Jan 22 20:59:52 i have 4 paints i cache and reuse so im not garbaging them over and over Jan 22 21:00:26 vs was suppsed to be "cs" but i typed in the cracks. cs is a canvas on the miniMap bitmap Jan 22 21:00:35 theblang: i don't.. and if you're using straight AsyncTask, that'd make a lot of things a huge PITA Jan 22 21:00:41 erase to invisible = no artifact. erase to black = artifact Jan 22 21:00:59 what i did was end up writing some wrapper code that allows me to make them static (tho i still keep them as inner classes just cuz) Jan 22 21:01:07 i do notice that if you write a pixel with an alpha of 0x00 the pixle is not written. thtas bad Jan 22 21:01:29 yeah, 0x00 is invisible Jan 22 21:01:51 the pixel should still be written, its only invisible at DRAW time Jan 22 21:02:10 agreed Jan 22 21:02:12 i.e. a pixel of 0x00112233 should be written into the bitmap but not displayed of course Jan 22 21:02:16 tophyr I see. yeah, I've got some activities with Async tasks, and Loaders. I want to transition to fragments but I am feeling a little overwhelmed. Thought it might help to make some of that stuff self contained maybe Jan 22 21:02:25 google got that part wrong but oh well :P Jan 22 21:02:38 I don't use AsyncTask, I just implement Runnable and use an ExecutorService Jan 22 21:03:35 yea asynctask was a pita. creating runnable's is much easier Jan 22 21:03:39 i keep meaning to open-source the stuff i have written for asynctasks and loaders.. ended up building a crap ton of stuff that makes it a lot easier Jan 22 21:04:09 I've also heard that unless you give the AsyncTask an executor it doesn't actually run on a background thread Jan 22 21:04:29 that's not true, they just run serially (in the background) by default Jan 22 21:04:44 there are two predefined executors, a serial and a parallel one Jan 22 21:04:50 or you can give it your own Jan 22 21:04:56 that's right Jan 22 21:05:06 whats a serial executor? Jan 22 21:05:11 if you queue up 5-6 asynctasks by default it will run them in order? Jan 22 21:05:20 I440r: a background thread that runs AT's in order one at a time Jan 22 21:05:20 ooooohhhhh !!! Jan 22 21:05:40 mental note on that one lol Jan 22 21:05:50 jonc-1: correct, they did that because too many people didn't understand that in a parallel executor, order is undefined Jan 22 21:05:55 that's bad, most of the threads I spawn will be io bound, so having serial threads makes no sense Jan 22 21:06:18 run them in the parallel executor then Jan 22 21:06:35 I just run them in a ExecutorService most of the time Jan 22 21:06:37 I've found much android happiness by writing code in (I call the push-pull style.) Basically, there's a single service runs background tasks; perhaps in multiple thread pools. This is the "push". Completed tasks post results to a global pub-sub queue; and anyone who cares can subscribe (typically on onResume(), and unsubscribe on onPause()) this is "the pull." By separating the originator of the task from who gets the completion s Jan 22 21:06:38 things work very nicely. Jan 22 21:06:42 typically using a thread pool Jan 22 21:06:54 yea this is def an alpha problem kbs, you called that one right Jan 22 21:06:59 but i srsly do not understand it Jan 22 21:08:30 the core framework is pretty ok for doing this sort of thing - between IntentService and Handlers, it's a very small amount of work to set up things like this; and most of the headaches of asynctasks/loaders are never needed Jan 22 21:09:09 kbs then you have to deal with parceling every result though, no? Jan 22 21:09:19 for what I am currently working on the only things needed on seperate threads are a bluetooth collector service, and IO Jan 22 21:09:34 or do you just ignore the binder part and stick em in a shared queue, is what you meant Jan 22 21:09:37 yea. if my bitmap is erased to black i get the artifact. if its not erased to black i dont Jan 22 21:09:40 tophyr: yep, exactly Jan 22 21:09:54 It's all in the same process; and the pub-sub can return results on the main thread as well via a Handler; so it's pretty simple to use Jan 22 21:10:03 makes sense Jan 22 21:10:17 iam using a preferencefragment./exit Jan 22 21:10:29 that "feels" ickier to me but i shy away from globals probably more strongly than i should Jan 22 21:10:38 to each their own Jan 22 21:10:46 I haven't found an overly nice way to avoid them in Android Jan 22 21:11:39 one coder's cache is another coder's global, or something like that :) Jan 22 21:12:06 haha true that Jan 22 21:12:20 I do like Observer / Observable Jan 22 21:12:30 globals are evil in the same way GOTO is evil. i.e. not at all Jan 22 21:12:42 goto is only evil when thats the only way of redirecting execution :P Jan 22 21:12:46 I'm trying to create a generic AsyncTask by reflection to consolidate rewritten code... does anyone know how may I refer to its generic types for parameters? Jan 22 21:13:11 Using DrawerLayout, my nav drawer is taking up the full width of the device when it is hardcoded to be only 240dp wide: http://pastie.org/8658256 what gives? Jan 22 21:13:32 ColinTheShots: you can't. type information like that goes away at runtime, it's only a compile-time thing Jan 22 21:13:40 <|0xD34D|> explodes: what dpi is your device? Jan 22 21:13:54 320 Jan 22 21:13:57 1200px wide Jan 22 21:14:02 (nexus 7) Jan 22 21:14:19 when i just have a ListView as the drawer, it works fine Jan 22 21:14:25 this compound view seems to confuse it... Jan 22 21:14:29 if I use a fragment transaction to remove fragments, does that also pop them from the backstack? Jan 22 21:14:31 at runtime the "type" is really just whatever the basest type that fits the expression.. ie, Something { T ob; } ---> ob is Object Jan 22 21:14:47 Something { T ob; } ---> ob is Foo Jan 22 21:15:22 tophyr: So it's not possible to create an AsyncTask that returns the parent type for a number of different asynctasks and then make it run callbacks for specific cases? Jan 22 21:15:28 actually, my second statement might not be correct.. might still actually be Object. can't figure out how it'd impl super, using my logic Jan 22 21:16:01 <|0xD34D|> explodes: have you tried setting the width of that RelativeLayout to nav_drawer_width and then the children inside that to match_parent? Jan 22 21:16:05 i'm not sure what you mean by "returns the parent type". i've done a lot with AT tho Jan 22 21:16:56 |0xD34D|: upon doing that, the subviews (NavProfileItemView and ListView) take up the full width of the device Jan 22 21:17:04 tophyr: normally I just skip them entirely, head to volley & gson. So much easier & faster Jan 22 21:18:07 Desarun: regarding generics? or parceling/sub-pub mechanisms? Jan 22 21:19:57 tophyr: so can I create an AsyncTask of and subclass it for each type I need to return? Jan 22 21:20:13 I was trying to handle it using reflection Jan 22 21:20:51 oh, sure. class MyAsyncTask extends AsyncTask { }, then class LeafAsyncTask { } Jan 22 21:21:08 ColinTheShots it would be ? extends ParentClass or something i think Jan 22 21:21:18 ahh Jan 22 21:21:35 but … the whole thing seems , odd ;) Jan 22 21:22:02 MyAsyncTask extends AsyncTask { } LeafAsyncTask { } might be what you're after? Jan 22 21:22:29 tophyr: Perhaps that's what I'm looking for Jan 22 21:22:45 replacing T with whatever expr you want.. T extends Parcelable, T super ThisParticularKindOfSprite, etc Jan 22 21:24:24 if I use a FragmentTransaction to remove fragments, does this also remove them from the back stack? Jan 22 21:25:01 nope Jan 22 21:26:17 i have multiple instances of widgets. when i hit a button, a preferencefragment opens - how can i distinguish the caller-appwidgetids within that fragment? Jan 22 21:26:33 getActivity.getIntend.getExtras(---) doens work Jan 22 21:26:53 it returns an appwidgetid which isnt among the widget-appwidget-ids Jan 22 21:27:14 tophyr: generics. Having had to modify a legacy production ASync system that used generics, we found it significantly quicker to use concrete type parameters. On the other hand we always knew what data we'd be getting from each source. Jan 22 21:27:27 |0xD34D|: doh; missing: android:layout_gravity="start" on the parent of the drawer Jan 22 21:28:25 Desarun: quicker to develop or execute? i can definitely see the advantages for development Jan 22 21:29:40 http://pastebin.com/GY9gdd8H Jan 22 21:30:06 I see a glimpse of the fragment that I want to be at the top of the back stack before the new fragment is added after I popBackStackx3 Jan 22 21:30:29 tophyr: Develop. As for execution, the only thing you can control is parse time. GSON outperformed the native Android json library by miles in internal testing Jan 22 21:30:38 but if I hit the back button once the BreathingStrategyFragment is shown, I get an error saying the DistressContainerFragment with view id XXXX cannot be found Jan 22 21:30:46 any ideas? Jan 22 21:30:52 that's what i figured Jan 22 21:30:56 Which is kinda strange considering they were both made by google.... Jan 22 21:31:01 i actually didn't even know android had an internal json library Jan 22 21:32:48 anyone here implement Search Suggestions? If so, did you use a CursorLoader to return a Cursor in your query() method? Jan 22 21:35:54 cigarshark: We've done that. If I remember rightly we held the result for whatever the user had just typed in memory, then discarded it (requesting new data) wether they addedd or removed a letter to the search Jan 22 21:36:42 Desarun, are you talking about the builtin SearchView widget, or your own search implementation? Jan 22 21:37:46 cigarshark: Our own. We couldn't get enough control over the look and feel using the widget. Wasn't that hard either, it's pretty much a textfield with delusions of grandeur Jan 22 21:38:53 haha. I read about lack of matching look and feel issue w/ the builtin, but its just so damn convenient. I'll stick with it for now since I'm a beginner at droid. Jan 22 21:42:40 I'm getting myself mixed up with what R is in android. I understand its created dynamically from resources in the project but... why cant I inspect it? e.g I can evaluate "R.id.customSeekBar" for example but why not R.id? (intellij) Something totally obvious about classes/types? Jan 22 21:44:16 rgr: So all those layouts you've got, you've specified ID's in some of the objects, right? Jan 22 21:46:08 I have one layout. I'm trying to understand how R works. See above. R.id.customseekbar cab be evaluated but not R.id. Maybe I'm not quite following the line of your query? Jan 22 21:48:28 Oh I think I have it. It's down to inner classes and static members. Jan 22 21:48:30 When you say evaluate, what do you mean? id is a static final class in R.java Jan 22 21:48:38 ^ Jan 22 21:48:50 heh. timing is everything. I get it now. Cheers. Jan 22 21:49:30 When I tried to "eval" R.id the IDE said nothing and I hadn't at that point opened R.java from the gen directory. I get it now. Jan 22 21:57:50 Could you explain to me what the android packaging does with the strings xml files from the values folder? I see I can break strings.xml down into 2 or more files and there is no need to change the app code - the layout xml file still uses "@string/mystring" - there is no need to specify the string xml file name from which the actual translation is done from. Jan 22 21:58:40 ok so i erase the minimap to alpha = 0x00 and i get no artifact yes?. Jan 22 21:58:54 well if i save that out and load it back in again it GAINS the artifact Jan 22 21:59:08 and the alpha = 0 areas are loaded in as black. not invisible Jan 22 21:59:19 so the PNG save is not saving the alpha channel Jan 22 22:01:23 rgr, file names in values/ don't matter.. It works at the resource level.. You tell it it's a string with Jan 22 22:05:28 Heya Jan 22 22:06:32 Anyone used this: https://code.google.com/p/java-universal-tween-engine/ Jan 22 22:06:39 Is it efficient? Jan 22 22:06:44 SimonVT: yes, I see that. But can you let me know the "mechanism" that the code extracts the strings? Are they all "precompiled" into something or is there another class created holding all the strings or does the runtime just grep through the xml? Jan 22 22:06:58 (just getting a feeling for all it how hangs together) Jan 22 22:07:18 hey there pilgrim Jan 22 22:08:44 How well supported is C++ in the NDK? Jan 22 22:09:29 what do you mean? Jan 22 22:10:10 the NDK uses the GNU devel tools including G++ Jan 22 22:10:38 personally why anyone would want to develop ANYTHING what so ever in c++ is beyond me lol Jan 22 22:11:21 yup its official. the PNG loading DITHERS the image it loads in Jan 22 22:11:49 b = BitmapFactory.decodeFile( Jan 22 22:11:49 retrorpg.FILE_PATH + "/minimap.png"); Jan 22 22:11:56 nick9998: Is the stdandard lib supported? Jan 22 22:11:59 this dithers the file it loads. how do i stop it doing that Jan 22 22:12:06 nick9998: And is C++11 supported? Jan 22 22:12:59 svm_invictvs:it uses g++, u can make a toolchain and check what version is it Jan 22 22:13:00 ahha BitmapFactory.Options.inDither = false Jan 22 22:13:01 rgr, XML resources are compiled into binary XML.. values/ is compiled into some resources.arsc file (if you unzip and apk), I'm not sure how it stores it there.. You'd have to check for yourself Jan 22 22:13:02 yay Jan 22 22:13:11 nick9998: hm Jan 22 22:13:27 nick9998: I've got a bunch of codew ritten in C++ I would rather not port to Java Jan 22 22:14:25 svm_invictvs:not a big c++ fan myself... Jan 22 22:15:00 nick9998: heh Jan 22 22:22:16 Trying to migrate my ant project to gradle. As an intermediate step, I'm considering using ant+ivy in my project to get dependencies Jan 22 22:22:24 does ivy support aars? Jan 22 22:22:29 i dont think so Jan 22 22:22:44 arg ok Jan 22 22:23:37 shigeru its pretty easy to convert especially if you dont have complicated build Jan 22 22:23:45 BitmapFact.Options o = new ....; o.inDither = false; <-- input is still being dithered Jan 22 22:23:48 like, custom tasks , etc Jan 22 22:23:54 hmm Jan 22 22:24:00 i also tried o.inScaled = false; Jan 22 22:24:07 im pretty intimidated x_x, i have a main ant project that depends on other ant projects Jan 22 22:24:12 my main ant project has an ndk component Jan 22 22:24:24 ah, ok. don't know about ndk Jan 22 22:24:26 i want to one by one convert the dependency ant projects into gradle projects Jan 22 22:24:34 without having to do it all at once you know? Jan 22 22:24:47 sure Jan 22 22:25:03 also everyone here is using eclipse so if i move one project to gradle i feel like it will make eclipse integration weird Jan 22 22:25:14 yeah ... Jan 22 22:26:40 the biggest impedance mismatch will be between the canonical gradle project structure of a /new/ project - eclipse wont like that -- but you can adjust gradle to use the oldish structure Jan 22 22:26:58 they make barrel connectors for that, right? Jan 22 22:27:24 let the eclipse devs keep building in eclipse, but they can use gradle from the command line i guess? you might have to point eclipse to the gradle cache so it can resolve jars or something meh Jan 22 22:28:04 i'm kinda surprised ADT hasn't gotten any gradle support yet; it seems dead Jan 22 22:28:25 not a good strategy imho; unless it was technically infeasible ;) Jan 22 22:30:14 rager i wouldn't bother with it yet if you are still using eclipse ;) Jan 22 22:30:22 oops, shigeru ^^^ Jan 22 22:31:00 thanks Jan 22 22:31:24 gradle is very flexible and can accomodate the eclipse - expected structure; you can just grab .jars and put them in libs/ and gradle can handle it Jan 22 22:31:32 g00s: is there any better IDE for android development? Jan 22 22:31:37 true Jan 22 22:31:45 IJ Jan 22 22:31:46 im worried that aars will complicate this Jan 22 22:32:08 x_x ill think about this later heading otu Jan 22 22:32:16 shigeru the tough part is making sure gradle itself handles your build system; there are still bugs which can be weird show stoppers. Jan 22 22:32:48 still seems weirdness with manifest merging of certain elements, etc Jan 22 22:33:13 I actually managed to switch the team over to Android Studio back when I was the whole team Jan 22 22:33:31 Cowabunga dudes. What OS do you prefer to run the Android SDK under? Jan 22 22:33:44 OS/2 Jan 22 22:33:52 SimonVT: thanks. Jan 22 22:33:53 hold on rager, let me find a torrent :P Jan 22 22:34:16 honestly, linux is still the best for that Jan 22 22:34:45 although note that values is xml too. Jan 22 22:34:51 I'm on a Mac, and I abhor running Java on it. Ubuntu inside a virtual machine would probably do just fine? Jan 22 22:34:58 don't do that. Jan 22 22:35:05 anecdotally i hear more people having problems with windows b/c of the usb drivers or some crap; followed by linux, and then hardly anyone complains about os x Jan 22 22:35:12 just run the dang shooting thing on OSX Jan 22 22:35:25 using modern java tools on a mac is indeed a pain Jan 22 22:35:30 it's almost linux, as far as you're probably concerned, anyway, with proprietary addons Jan 22 22:35:37 I've never had a single problem with java on this mac Jan 22 22:35:41 seems little point in using AS when you get pretty much the whole shebang with more flexibility in intellij CE. Jan 22 22:35:48 my team as well as most of the google team (as far as i understand) use os x Jan 22 22:35:53 ekx what version of mac though? the oracle java is pretty sane :| apple bundled java was a clusterfuck nightmare Jan 22 22:35:58 if your team uses OSX Jan 22 22:36:00 then use OSX Jan 22 22:36:21 that way, you don't end up cursing [arbitrary member of your team] when things don't work for *you* at 3am, but everybody else is doing fine Jan 22 22:36:34 rager: you're trolling or joking I assume. OSX hasnt got anywhere near the flexibility of Linux for Android development and development in general unless its for an iOS device or Mac. Jan 22 22:36:35 g00s: I'm using the latest OSX, but I don't run Java because I've been burned by lax security (and I still remember the pain) Jan 22 22:36:56 ekx lax security? do you have th browser plugin enabled or some crap? Jan 22 22:37:04 g00s: that's the biggest problem, yeah Jan 22 22:37:11 ekx, you're agonizing over pointless details Jan 22 22:37:13 just disable the plugin :| Jan 22 22:37:36 rgr: clarify? i use osx daily and development is 100% of my job, for android as well as ios Jan 22 22:37:43 he's gone Jan 22 22:37:44 even have a vm set up and work on asp.net fairly regularly. Jan 22 22:37:56 but you don't have the flexibility in OSX for doing dev Jan 22 22:37:57 well then Jan 22 22:38:14 you do have *most* of it, though Jan 22 22:38:26 * tophyr scratch head Jan 22 22:38:28 what flexibility? Jan 22 22:38:36 * g00s scratches head too Jan 22 22:38:45 i guess i obviously don't need whatever extra flex linux offers then Jan 22 22:38:50 you probably don't Jan 22 22:39:03 I've had headaches dealing with native builds in OSX, but that's about it Jan 22 22:39:05 i do have a linux VM set up for aosp/cm builds, but that's mainly because those are set up specifically for linux Jan 22 22:39:12 everything else runs in its own little world Jan 22 22:39:21 bah I'll install Ubuntu in a VM for now Jan 22 22:39:25 lol Jan 22 22:39:31 and not just run it natively? that's silly. Jan 22 22:39:38 I'll change my mind later, I'm sure Jan 22 22:39:50 that's got all the headaches of linux and OSX, combined! Jan 22 22:39:55 yay! Jan 22 22:39:58 plus extras from virtualization Jan 22 22:40:25 my fresh OSX install is kernel panicking, so we'll see how disastrous this becomes Jan 22 22:40:34 10.9? Jan 22 22:40:41 ekx makes is like more difficult hoping to make his life less difficult :) Jan 22 22:40:53 *sigh* yes Jan 22 22:40:59 at least it's not 10.8 Jan 22 22:41:00 sure sign of a coder Jan 22 22:41:05 ugh Jan 22 22:41:07 ekx kernel panic with vbox ? Jan 22 22:41:16 g00s: no, native OS X has sleep problems Jan 22 22:41:29 hm. i still use 10.7 Jan 22 22:41:31 :) Jan 22 22:41:51 seems like both apple and google require lots of users for beta testing their products these days :) Jan 22 22:42:05 * g00s hasnt signed up for the beta testing program ;) Jan 22 22:42:05 10.8 was a steaming pile, and it came installed on my work mac Jan 22 22:42:25 so it took upgrading to 10.9 to make the laptop fully usable Jan 22 22:42:32 *never* buy the cheapest mac Jan 22 22:42:39 hah, i remember holding off on 10.8 because ethernet port didn't work on my modem mbp, and then i totally forgot about it Jan 22 22:42:40 yeah 10.9 for me has been solid as a rock Jan 22 22:43:12 but then, the loadout IT gives our dev machines is as powerful as they can make them Jan 22 22:43:16 i'll try it when 10.9.2 comes out Jan 22 22:44:20 ah capella is here Jan 22 22:44:27 * g00s eagerly awaiting FF 27 with TLS 1.2 Jan 22 22:44:31 \o/ Party Jan 22 22:44:56 capella wtf, doesn't chrome use NSS ? Jan 22 22:45:04 * capella just got his first Metro project into Nightly Jan 22 22:45:19 NSS? Of course! (I thought??) Jan 22 22:45:27 afaik! Jan 22 22:45:56 ekx, just dual boot your Mac to run Linux Mint and OS X. I dual-booted by Mac Mini to run either OS using the reFit app. Took ten minutes to figure out. Jan 22 22:46:06 cigarshark: cool, probably next time Jan 22 22:47:03 Linux Mint runs way faster than Mavericks on the Mac Mini. I think Mavericks is bloatware, but it goes unnoticed because most Macs have great hardware specs. Jan 22 22:52:03 so this module i just imported requires "android-support-v4" as does my app. Jan 22 22:52:05 the ubuntu download took so long I already set up Android in OS X Jan 22 22:52:17 doh Jan 22 22:52:21 it includes its own copy, and that means i get the error: Multiple dex files define Landroid/support/v4/accessibilityservice/AccessibilityServiceInfoCompat$AccessibilityServiceInfoVersionImpl Jan 22 22:52:50 how do i configure gradle to use the same jar as my app? Jan 22 22:53:11 cigarshark, mint = ubuntu = debian = a POS that cant even follow the FHS. Jan 22 22:54:44 er, how do i configure the module's gradle to use the same jar as my app? if i include the same dependencies line, i get a different set of errors around the idea that "package android.support.v4.view does not exist" Jan 22 22:54:54 I440r: what is FHS? Jan 22 22:55:24 filesystem heirachy standard Jan 22 22:55:33 what files belong in what directories on the file system Jan 22 22:55:45 i.e. Terminfo does NOT belong in /lib/terminfo Jan 22 22:55:49 not now, not ever Jan 22 22:56:23 I have all the various OS' at home. Mac Mini dual booted with Mint 16, Macbook Air, dual booted Win 8.1 / Mint 16, and Win 7 dual booted w/ Min 16. Jan 22 22:57:00 Gentoo FTW (/me ducks) Jan 22 22:57:07 so I can pick and choose. My favorite amongst all these is Mint 16. I hated all Linux flavors until I tried Mint for the first time. Jan 22 22:57:20 Alright, how's it going? Jan 22 22:57:21 if I need Context within a fragment is calling this.getActivity() the proper way to retrieve it? Jan 22 22:57:53 and all my dev tools run extremely well on Mint. I've got everything I need except for Adobe products. Jan 22 22:58:36 the FHS is not anything that matters. Once you know where shit goes, you're done. Jan 22 23:00:13 except when you have to add duct tape code to check "is it here... or here... or here... or here" for the various assinine distributions Jan 22 23:00:20 instead of "its here.. i know it" Jan 22 23:03:25 why the fuck cant these modules share the same library Jan 22 23:04:06 fucking android has a steep learning curve Jan 22 23:04:33 yes Jan 22 23:04:34 i guess thats good for employment purposes Jan 22 23:04:53 and android studio does too, but it is so volatile between versions you cant search for any help Jan 22 23:05:10 sheesh Jan 22 23:05:24 yeah, you basically have to hack your way through Android Studio. Very little info and slow going at first. Jan 22 23:06:01 coming from Eclipse, it was a real pain in the ass. Intellij people would probably pick it up quicker. Jan 22 23:06:19 +1 cigarshark Jan 22 23:06:24 i was a goddamn wizard in eclipse Jan 22 23:06:29 holy fuuuuuuuck Jan 22 23:06:38 i can't search for help at all on this Jan 22 23:06:43 now i'm poudning my head bloody with AS/gradle Jan 22 23:06:53 me too tophyr Jan 22 23:06:54 me too Jan 22 23:07:39 gradle seems to me, like the looked at make, and said "40 years of the same thing is enough. i want to build that, but better". Jan 22 23:07:49 so they set out to build a better grocery getter Jan 22 23:07:49 Darkula theme rocks Jan 22 23:07:55 and they built fucking SpaceX Jan 22 23:08:00 in order to get fucking groceries Jan 22 23:08:04 yeah, Gradle is the fuckin wild west. Jan 22 23:08:23 not too many ways to fuck up Maven. it was limited and that wasn't such a bad thing. Jan 22 23:09:23 its like moving back to Ant kinda. everyone has a different build file w/ all kinds of hacks to research before you understand it. Jan 22 23:09:24 its worth the pain Jan 22 23:09:35 learn AS/IDEA Jan 22 23:09:47 it puts eclipse to shame on EVERY level Jan 22 23:09:55 but yeah, Maven was getting boring. keeps things interesting to have this new shit. Jan 22 23:10:47 I am liking AS now. Jan 22 23:10:58 So, i added this library module that depends on android-support-v4 Jan 22 23:11:02 my app also depends on this jar. Jan 22 23:11:12 i like AS, aside from its bugs.. its gradle i'm hating on Jan 22 23:11:41 how do you guys typically work with AS/gradle apps that depend on library projects? Do you load the main project + the library projects into AS and work from there? Or do you set up an internal artifact repo and have dependencies pushed/pulled to/from there from the main project? Jan 22 23:11:44 osmdfsdkfjslkdfj Fuck Jan 22 23:11:53 Nevermind, gotta type all this out Jan 22 23:11:56 elsewhere Jan 22 23:11:59 I added my library project as a Module Jan 22 23:12:08 Go through the Add Module wizard in AS Jan 22 23:12:15 how do snapshot dependencies relate to that Jan 22 23:12:21 is that a completely seperate thing Jan 22 23:12:50 Its not working. Jan 22 23:12:51 snapshot deps are just a versioning mechanism from maven Jan 22 23:12:56 shigeru: 1st, for me. if i'm not pulling it from maven i'm building it locally Jan 22 23:13:12 basically, the snapshot is an unreleased version of a jar that is subject to change Jan 22 23:13:12 Where is the add module wizard, i want to make sure we're talking about the same thing, shigeru Jan 22 23:13:20 File->Add Module Jan 22 23:13:30 sorry New Module Jan 22 23:13:47 also why can't i delete this folder from AS Jan 22 23:13:48 shigeru: I generally just stay away from dependencies I can't add from build.gradle (a la maven) Jan 22 23:13:55 I think i get what you mean cigarshark Jan 22 23:14:04 it has a module wizard to add Android Library to your project Jan 22 23:14:15 but when I can't, I just drop the library into my "libraries" folder at the root of the project and make a build.gradle for it Jan 22 23:14:16 what if i want to share those dependencies with other teams Jan 22 23:14:18 or open source them? Jan 22 23:14:39 do you still load them as AS modules Jan 22 23:14:49 I would create the library through the module wizard or else AS won't treat it properly Jan 22 23:15:16 my build.gradle file tells the people what dependencies they need to hunt down if a series of nuclear strikes took down the maven repositories Jan 22 23:15:23 shigeru: the only real advantage to hosting an artifact repo is not having to recompile the code all the time, as far as i can tell Jan 22 23:15:25 you can't just do it all through gradle or your IDE won't behave like you want it to Jan 22 23:15:30 shigeru that wizard is to actually code up a module, not import code someone else wrote Jan 22 23:15:42 ohhhh, sorry Jan 22 23:15:55 u have a lib that is already built Jan 22 23:15:58 yeah Jan 22 23:16:10 multiple teams within my company contribute to that lib Jan 22 23:16:12 then add the compile-time dependency in build.gradle Jan 22 23:16:13 and its shared between two apps Jan 22 23:16:20 and... oh Jan 22 23:16:27 not built, i have to source code for it Jan 22 23:16:36 do you guys have your own maven repo out there? else, i'd submodule it. Jan 22 23:16:36 it has a build.gradle tho Jan 22 23:16:47 i *could* load it as a library module in AS with the main project Jan 22 23:16:56 or it seems like i could host a maven repo and pull it from there Jan 22 23:17:13 is it possible to disable an irq handler created by another process Jan 22 23:17:14 we dont really have anything set up now though, no maven repo, we're still using ant Jan 22 23:17:19 we want to transition to gradle/AS though Jan 22 23:17:31 you could just use git submodules if you don't expect to edit the lib from the app projects Jan 22 23:17:39 Ok i'm just ask the about my real problem: I have someones repo: https://github.com/umano/AndroidSlidingUpPanel Jan 22 23:17:47 how do i import that shit into my project? :/ Jan 22 23:17:53 rager thats true, submodules sounds like it could work Jan 22 23:18:06 blocky: IRQ as in interrupt request? i didn't even know those were accessible in android. Jan 22 23:18:07 why dont more people suggest that instead of pull from an artifact repo? Jan 22 23:18:26 submodules are a PITA Jan 22 23:18:29 and not the bread, man Jan 22 23:18:34 lol Jan 22 23:18:36 submodules are easy as pie Jan 22 23:18:49 we currently have a submodule heavy set up Jan 22 23:18:50 okay, I retract that statement Jan 22 23:18:54 with submodules depending on submodules Jan 22 23:18:57 submodules are a PITA when your team doesn't know how to use git Jan 22 23:18:57 its kind of hairy Jan 22 23:19:08 and my team really doesn't even know how to use git, beyond whatever GUI they've got Jan 22 23:19:12 rager: that i can def agree with Jan 22 23:19:22 Here is what I'm proposing i do, not that anyone is listening: Jan 22 23:19:27 our main repo has like 4 levels of submodules -_- Jan 22 23:19:32 explodes: have you tried just using the maven repo? Jan 22 23:19:36 it seems to be in there: https://github.com/umano/AndroidSlidingUpPanel/blob/master/pom.xml Jan 22 23:19:36 0_0 Jan 22 23:19:42 if its just a jar, you could certainly build the jar as a snapshot version and put it in an in-house maven repo Jan 22 23:19:48 what is maven :I Jan 22 23:19:53 then just declare it in gradle Jan 22 23:19:57 ugh Jan 22 23:20:24 and press the 'sync project with gradle' button Jan 22 23:20:31 cigarshark do you recommend that over using submodules? Jan 22 23:20:42 rager: how do i add a maven module? Jan 22 23:20:42 submodules seems easier to me Jan 22 23:21:01 if you have to debug into the 3rd party lib for some reason, you could set a breakpoint in their source code Jan 22 23:21:11 I'd tell you exactly what to do, but then you're stuck having to ask people every time Jan 22 23:21:12 http://www.gradle.org/docs/current/userguide/artifact_dependencies_tutorial.html Jan 22 23:21:13 with a jar file, that isn't going to work in AS Jan 22 23:21:17 explodes: see that link Jan 22 23:21:36 im fine not beign able to step into jar source code Jan 22 23:21:53 is an 'Android Library' just a jar file? Jan 22 23:22:01 i have aars too Jan 22 23:22:08 shigeru: You trust other devs? Jan 22 23:22:14 cigarshark: Yes. Jan 22 23:22:20 desarun yeah i guess Jan 22 23:22:24 (lazy) Jan 22 23:22:30 well , check it out Jan 22 23:22:45 cigarshark: But if you've got Android Library projects that your main project uses, make damn sure they've all got the same version of the library. Jan 22 23:22:45 just create the JAR file, and copy it to your local maven repository Jan 22 23:22:56 and then declare it in build.gradle Jan 22 23:23:10 dont' even bother setting up your own maven repository over HTTP or whatever Jan 22 23:23:22 cigarshark: Sorry, I thought you meant Android Support Library. Jan 22 23:23:24 just copy it manually down in your /.m2 folder Jan 22 23:23:45 see, I am new to droid myself. I know a shitload about maven and a little bit about gradle and AS. Jan 22 23:23:57 Desarun cigarshark thats another thing i was wondering. if i go the submodule route, if i have project A depending on project B which depends on project C and A also depends on C, and there ends up being different versions of C in those repos, what happens Jan 22 23:23:59 now android studio is automatically re-adding files i am trying to delete Jan 22 23:24:12 An Android Library is normally a project that comes with resources, emaning it has its own R.*jazz* file, meaning you can't bundle it into a jar Jan 22 23:24:20 i think i read that maven (and hopefully gradle?) can resolve those kinds of dependency conflicts Jan 22 23:24:54 who controls project C in your scenario? is code for C changing? Jan 22 23:25:07 cigarsharki could copy stuff into my local /.m2 but wont that be a pain for other team members Jan 22 23:25:15 ok rager: how do i take pom.xml and make it a "compile foo" dependency entry :I Jan 22 23:25:24 i control A B and C cigarshark Jan 22 23:25:44 the only way to ensure you have the same code /jar is to release the jar. hint: use the mvn release plugin. If the jar is not a snapshot version, it means the code is FIXED. Jan 22 23:25:46 C changes, and the submodule for it is updated in project A but not B Jan 22 23:26:14 when you see SNAPSHOT, it means you don't really know whats in the jar. Jan 22 23:26:50 sometimes i want to cry Jan 22 23:26:57 not often Jan 22 23:26:59 but sometimes. Jan 22 23:27:28 So our jars are built using Jenkins & SVN. When it's built, it checks the compiled jar back into SVN appending the revision number. Not fancy, but it does work. Jan 22 23:27:29 however, maven will automatically update new snapshot jars. It tries to download the latest snapshot jar when you do a build. Jan 22 23:28:25 is this a team environment or are you the only one working on A, B, and C? Jan 22 23:28:38 are snapshot jars pretty much only used with your local m2 repo when iterating on something before checking it in/ Jan 22 23:28:44 team environment Jan 22 23:29:23 yeah, snapshot jars are used for ongoing development and they contain code that is in flux. Jan 22 23:30:01 a developer can change some code and deploy it as a snapshot jar. then this snapshot jar will be downloaded when another team member runs a build. Jan 22 23:30:27 a release jar is cached once and never downloaded by maven again Jan 22 23:31:03 if the jar doesn't say SNAPSHOT, then there should only be one version of that jar, and it should never change. thats the convention. Jan 22 23:32:16 so you shouldn't have more than one version of a project-1.0.jar, but project-1.0-SNAPSHOT.jar could be changing all the time. Jan 22 23:33:12 in the build.gradle/pom.xml file that gets checked in, does that usually depend on a snapshot artifact or one with a fixed version? Jan 22 23:33:23 checked in to your teams version control system Jan 22 23:34:15 i check in poms with snapshot in them all the time Jan 22 23:34:40 Is there any way I can write to settings of my app, without launching the app? Jan 22 23:34:42 I mean like the IDE Jan 22 23:35:04 if you are doing an official release of a product, you make sure that all of its dependencies are NOT snapshots. then you check in a POM without snapshots in it, and you can tag the release. Jan 22 23:35:05 to reach your SharedPreferences Jan 22 23:35:31 Because sometimes you want to toggle some rarerly set setting, such as isAuthorized, or isSetup, or isFirstRun, etc Jan 22 23:35:37 its just to make sure that any jars used by the project have an official, unchanging version contained in the repository, or tagged in SVN Jan 22 23:37:09 smallfoot-: looking online, have a look at this in the shell Jan 22 23:37:09 data/data/your.app.package.name/shared_prefs Jan 22 23:37:48 oh man i really need a coffee, i just did != instead of == TWICE!... no wonder my code wasn't working... *sighs* Jan 22 23:38:21 smallfoot-: And as the saying goes... Got root? Jan 22 23:40:57 Desarun, I don't know. I am talking about running the Android image from Android Studio in VM Jan 22 23:42:44 smallfoot-: can you be a bit more specific? Sounds like Genymotion. When you type 'adb devices -l' in a terminal on the host, can you see the device? Jan 22 23:44:23 Hello all. I'm trying to get this code working(MainActivity.java http://pastebin.com/6NPYmBq8 MyLocation.java http://pastebin.com/5XLFRkcd ) but I cannot get it working for the life of me. I get this error http://pastebin.com/nvrV9YgK Jan 22 23:46:51 Desarun, no not Genymotion, I don't use proprietary software. I use Android Studio and use its built-in Android images and emulator, I believe it uses QEMU Jan 22 23:47:16 Desarun, I have no device plugged in. I only use Android Studio which runs Android in a VM Jan 22 23:50:01 <|0xD34D|> rOOb: that's an odd error and I noticed xposed in the stacktrace so perhaps you have some xposed module causing you a problem Jan 22 23:52:39 |0xD34D| Hmm. I do have xposed installed but I dont think I'm using any modules from it. Jan 22 23:53:31 rOOb: see, i think it is your asynctask and the call to the progress dialog Jan 22 23:53:46 Looks like I had XblastTools module. Just disabled it and am rebooting phone now Jan 22 23:53:54 okay Jan 22 23:53:59 smallfoot-: So the link above I gave you makes me believe that your SharedPreferences are all stored in an XML file on the VM's file system. I'm not sure how you get this file out in your specific case, but normally you'd use ADB to 'pull' the file out via a shell, edit it, then 'push' it back on, which would also normally require root Jan 22 23:54:01 burmat It seems like that but I have no idea why or how to fix it Jan 22 23:54:09 I've been banging my head for days Jan 22 23:54:31 yeah, it doesn't stand out to me either. i also don't typically use progress dialogs though, so maybe i am not the best person to ask Jan 22 23:54:47 Desarun, I see Jan 22 23:55:06 Desarun, I wish there was a easier way... Jan 22 23:55:07 burmat Is there perhaps a better way to show the user something is happening? Jan 22 23:55:19 I was under the impression progressdialogs is what I want to use. Jan 22 23:55:22 smallfoot-: Buy a damn cheap Nexus S.... Jan 22 23:55:46 Desarun, Why? What would that be good for? Jan 22 23:56:12 Well you could root it, then you wouldn't have to figure out how to connect an ADB session to your VM. Jan 22 23:56:20 rOOb: sure, that is an obvious thing to use, but maybe just take it out for now to see if that is triggering the crash so you know to look into the docs a little deeper.. Jan 22 23:58:18 Desarun, after I root it. How would I change the variable? With a file manager and a text editor, and edit the XML file by hand? Jan 22 23:58:39 Yup. Jan 22 23:59:32 I've done it with whole databases before. Well, a table editor & not a text editor. Jan 22 23:59:44 What is a table editor? Jan 23 00:01:31 Something that helps you edit SQL tables, inside of a database. Jan 23 00:02:00 Oh Jan 23 00:02:20 Yeah, SharedPreference might save in a SQLite database... I don't know how SharedPreferene stores configs Jan 23 00:02:50 smallfoot-: a file Jan 23 00:02:54 like a text file Jan 23 00:03:02 Do you know any good table editors? and do you know any good text/code/ide editors? Jan 23 00:03:05 for Android Jan 23 00:03:35 as in an app or something Jan 23 00:03:42 or are you talking on a computer side Jan 23 00:04:58 yeah, idk for android. I used to have a good text editor app but not I just do everything through adb manually Jan 23 00:05:16 now* i just. I don't remember the name of the app. Jan 23 00:06:56 smallfoot-: If you look at the link I gave you above, it says that SharedPrefs are stored in XML in key-value pairs. Jan 23 00:07:04 burmat, Android app Jan 23 00:07:19 Desarun, I see Jan 23 00:07:46 where is that link? Jan 23 00:09:01 ahhh, Desarun is right. Sorry I mislead you. I forgot that Jan 23 00:09:29 I tell a lie, I didn't give you any link Jan 23 00:09:29 http://stackoverflow.com/questions/1243079/how-to-examine-sharedpreferences-from-adb-shell Jan 23 00:10:09 Hah, someone asked how to edit it via ADB less than two minutes ago.... Was that you perchance? Jan 23 00:11:10 smallfoot-: well herro there "Fred" lol Jan 23 00:26:44 how can I specifiy an imageview to be sized relative to the accompanying/neighbour text field? ie standard need to make a small label icon about the same size as a capital "W" for example. em measurements? Jan 23 00:26:46 burmat, hey, how do you know thats me? Jan 23 00:27:12 oh, you saw me post that comment! lol Jan 23 00:27:22 yeah, thats me! Jan 23 00:27:47 but then again, Fred really isn't my name, I stole it from Fred Flintstone :D Jan 23 00:29:25 hahaha jeeze Jan 23 00:29:57 :D Jan 23 00:30:26 Desarun, yes it was me! :D Jan 23 00:31:06 that's funny. Jan 23 00:33:30 rgr: a shot in the dark, but does using compound drawables get you there? Jan 23 00:45:37 I have a webview loading a url. and I have a button that calls a method to load another url to the same webview. For some reason it doesn't find the file. Is this a problem with the way I'm loading the url? Jan 23 00:48:14 yes Jan 23 00:49:55 how do I load it? durka42 Jan 23 00:50:15 well generally webView.loadUrl Jan 23 00:50:20 but I would guess you tried that and it didn't work Jan 23 00:50:23 you didn't say how you tried to do it Jan 23 00:51:00 1 sec let me upload code Jan 23 00:53:21 http://goo.gl/JYQf9s Jan 23 00:55:09 relativelayout always screws with me :( Jan 23 00:55:15 JakeWharton: hey, i've got another Spoon question for ya if you're around Jan 23 00:55:32 durka42, http://goo.gl/JYQf9s Jan 23 00:55:44 tophyr: go Jan 23 00:56:21 g00s: I'd have suspected you'd have written some sort of constraint/rule based layoutmanager by now :-) Jan 23 00:56:37 hmm sphynx_cat is that the right way to reference an asset Jan 23 00:56:38 https://stackoverflow.com/questions/8474821/how-to-get-the-android-path-string-to-a-file-on-assets-folder Jan 23 00:57:40 i'm going through all my build's tests and such since i finally got permission to spend a few weeks just being a build engineer/sdet/whatever, getting automation going awesome. i'm playing with both spoon and the stuff provided by android tools - it looks like connectedInstrumentTest, provided by android tools, is the same basic idea as Spoon: run all the tests on all connected devices, concurrently Jan 23 00:58:56 have you played with that at all? i'm wondering if spoon provides anything further than that (other than a gorgeous results display). i'd like to start tinkering on one or the other, to allow whichever tool to start emulators and run through a list, instead of being limited to "all connected devices" Jan 23 00:59:10 sphynx_cat: actually, I saw somewhere else that android_asset should work for the webview Jan 23 00:59:18 but, did you leave out the "webcontent/"? Jan 23 00:59:23 or maybe it can't directly load a gif? Jan 23 00:59:41 that is what I'm currently thinking. Maybe it can't load a gif. Jan 23 00:59:59 i looked at spoon-runner and it seems fairly straightforward how it connects to the device list and spawns a thread for each.. haven't seen android-tools but i imagine it's similar.. really, (and without downplaying spoon at all) i'm wondering if it provides anything further than what the new android-tools do Jan 23 01:00:58 durka42, found this http://stackoverflow.com/questions/15733749/read-gif-images-using-webview-android I'll let you know once I implement it to see if it works Jan 23 01:05:51 JakeWharton: hope i didn't come off as rude.. i think spoon's the shit, i'm just wondering which tool i should start hacking on Jan 23 01:06:10 i got distracted. let me read Jan 23 01:06:23 kk, thx Jan 23 01:06:46 ouch. how the hell to force a rebuild of R.java in intellij. Jan 23 01:06:54 build > rebuild project Jan 23 01:07:30 tophyr: spoon adds screenshots, gifs, grouping the results of one test across all devices, TV display, and the high-level initial dashboard Jan 23 01:07:37 it was built before the new Gradle-based stuff Jan 23 01:08:11 nobody is working on it now, but we had planned to circle back and integrate much more tightly with the new stuff and add things like the new video recording APIs and native screenshot APIs Jan 23 01:08:40 ok, cool, that's what i thought. video would be badass Jan 23 01:09:37 would there be any interest from you guys in being able to specify a list of emulators to run, max X emus at once? for a large test matrix, obviously you can't have 20+ emulators all running at the same time, but a pool of 2-4 cycling through the list would go quite well Jan 23 01:10:09 that's what i'd *like* to build, once i get myself familiar with how all these tools work underneath Jan 23 01:12:58 durka42, I tried this but it didn't work. http://goo.gl/yWu552 Jan 23 01:13:06 could I be sourcing the image wrong? Jan 23 01:13:13 what's the error? Jan 23 01:13:27 same as before. web page not available Jan 23 01:13:42 web page at ...blinking_wolf.html could not be loaded Jan 23 01:13:49 as the requested file was not found Jan 23 01:14:28 maybe you got the path wrong Jan 23 01:14:39 I see the directories are different in onCreate vs. load_monster Jan 23 01:14:52 you're genius Jan 23 01:15:02 * sphynx_cat slaps forhead Jan 23 01:15:14 d'oh Jan 23 01:15:27 nice!!!!!! it works but the gif isn't animated Jan 23 01:22:16 possibly some intellij whiz can hold my hand. I got in a bit of a mess renaming a project. Almost there I think but I cant get R to be regenerated and the main.xml wont show a layout preview saying "no render target was chosen". I cant find anything (that I deem) relevant to configure or any helpers in google. pointers? Jan 23 01:35:17 Is it possible to generate swipeable tabs on the fly? Jan 23 01:41:37 rgr: have you tried re-importing your project? Jan 23 01:41:51 and are you managing it with gradle or just Intellij/AS Jan 23 01:42:19 R0b0t1, pretty sure it's possible to create any UI stuff on the fly Jan 23 01:42:39 define:on the fly Jan 23 01:44:01 when your user has tossed the phone into the air, but before it hits the ground Jan 23 01:44:31 context.inflate(parachute) Jan 23 01:44:53 durka42: there's an app for that Jan 23 01:45:05 a freefall detector? Jan 23 01:45:08 wouldn't be too hard Jan 23 01:45:14 https://play.google.com/store/apps/details?id=com.bytemods.throw_phone&hl=en Jan 23 01:45:22 https://play.google.com/store/apps/details?id=com.carrotpop.www.smth&hl=en Jan 23 01:45:35 like.... best games ever? Jan 23 01:45:51 oh yeah there was that one game about how high can you throw your phone Jan 23 01:46:00 and apple banned it for encouraging people to damage their devices Jan 23 01:46:04 or did google ban it for that reason Jan 23 01:46:05 heh Jan 23 01:46:06 can't remember Jan 23 01:46:11 well, since those are play store links.... Jan 23 01:47:11 http://hexus.net/mobile/news/general/58613-iphone-hurling-sports-game-app-banned-apple-app-store/ Jan 23 01:48:52 Hello all. I have a progressdialog inside a AsyncTask that does seem to be working right. Its not displayed. Anyone have any ideas? I think its related to the MyLocation2() stuff but I dont know what to change. All the code is on here. http://pastebin.com/hHT5ykwN Jan 23 01:50:50 rager: Well I was not sure if I would be able to modify menu text for like drop downs once I pushed the project to the phone Jan 23 01:51:57 say I create a drop-down view or whatever it is called (sorry) Jan 23 01:52:09 can I change what is in the dropdown based on what my user is doing? Jan 23 01:53:11 you can do everything Jan 23 01:53:17 the limit is how hard it is to do the things Jan 23 01:53:28 ... Jan 23 01:53:46 dropdown/spinner views are generally AdapterView subclasses Jan 23 01:54:05 those things are backed by datasets you can freely edit and have displayed as soon as they're changed as long as you tell the data adapter to do so Jan 23 01:54:35 like... look at some examples for ListView, at least Jan 23 01:55:30 Alrighty Jan 23 01:55:34 so I can know what to assume, what is your current level of knowledge on the android api's? Jan 23 01:58:15 actually - I'm headed out soon Jan 23 01:58:21 Right. Almost there. One head scratching now. In intellij in my mainclass.java the line 'public class MainActivity extends Activity {' shows that MainActivity is "not used". its referenced in the manifest. Where else should it be? Jan 23 01:58:31 rager: Pretty low Jan 23 01:58:45 MainActivity.java that should read Jan 23 01:59:09 but so I think I figured it out - the xml menu is turned into Java, and I could theoretically skip the xml and autogeneration to have polymorphic menus? Jan 23 01:59:41 speak less vaguely Jan 23 01:59:50 what are these menus you're making? Jan 23 01:59:55 and how is the user supposed to interact with them? Jan 23 02:00:09 ListView does not allow the definition of its children in XML, for instance Jan 23 02:00:28 and I'd guess that it's the backbone of almost any app Jan 23 02:01:05 the ListView is populated using a ListAdapter or something similar Jan 23 02:01:29 the ListAdapter knows how many rows to make and what to put in them based on what data is backing it and what the defined behaviors are for creating new children Jan 23 02:01:46 you are free to assign different datasets and different adapters to the listview, even, to change what the listview displays Jan 23 02:02:00 and you're also able to add filters to the listview to alter the members of the backing dataset that are even displayed Jan 23 02:02:20 Well, perhaps it's only fetching the string from the menu resources Jan 23 02:02:21 yeha Jan 23 02:03:05 rager, well, really I guess I am confused about why resources are managed the way they are Jan 23 02:03:13 explain Jan 23 02:03:43 I see where the adaper is made but I am slightly confused as to why the menu text is stored the way it is (in the resource class, which is apparently autogenerated from xml) Jan 23 02:05:34 you mean R? Jan 23 02:05:48 R.java is a way to refer to the many non-code resources in your project from java Jan 23 02:06:18 it seems neater than having to load and track all the resources manually, at least Jan 23 02:06:26 but it's weird, yeah Jan 23 02:07:50 So... how do I specify a layout besides in an .xml file? Jan 23 02:07:59 you build it programmatically Jan 23 02:08:11 http://developer.android.com/guide/topics/ui/declaring-layout.html < dox Jan 23 02:13:10 I just don't understand why data is partitioned this way Jan 23 02:13:14 it is still making no sense Jan 23 02:13:36 but I guess if I just don't autogenerate anything I'll see if I get something working Jan 23 02:17:40 it's split up because data is data Jan 23 02:17:42 and code is code Jan 23 02:17:57 if you keep your layouts separate from code, they're more managable Jan 23 02:19:00 would be cool if java got the ?. operator; i guess i was dropped from java 7 coin :( Jan 23 02:20:01 rager: It's possible to separate data in such a way that doesn't prevent you from doing things you may want to do at some point in the future, like not use static resources Jan 23 02:20:05 is that the null-check-and-member-access operator? Jan 23 02:21:19 durka42 yeah Jan 23 02:21:49 yeah that's a useful one Jan 23 02:21:59 if only java had macros… :p Jan 23 02:22:00 that would make coding at least... 30% more pretty Jan 23 02:22:12 I mean, the null-check operator Jan 23 02:22:15 not macros Jan 23 02:22:24 I mean lisp macros of course, not C macros Jan 23 02:23:02 oh, that's a different story, lol Jan 23 02:23:34 anyone noticed http://www.android.com/kitkat/ Jan 23 02:23:36 is down? Jan 23 02:23:41 or just for me.. Jan 23 02:23:48 's just you Jan 23 02:23:57 404. That’s an error. The requested URL /kitkat/ was not found on this server. That’s all we know. Jan 23 02:24:37 just you Jan 23 02:25:13 android.com works.. Jan 23 02:25:16 only kitkat not -.- Jan 23 02:25:17 -.- Jan 23 02:25:57 that's weird.. if i access android.com i still get some articles about 4.3... Jan 23 02:26:13 FabianH: that's how they 'give you a break' Jan 23 02:26:21 I think you have a time traveling internet connection Jan 23 02:26:28 yes Jan 23 02:26:37 would be cool :D Jan 23 02:27:57 and coincidentally, -.- is "k" Jan 23 02:28:55 ? Jan 23 02:29:17 * kbs is an old-fart ham and sees morse code in random situations. Jan 23 02:30:18 I keep hoping one of these days someone will honk in morse and say something interesing. Jan 23 02:32:05 Why did Google name 4.4 "KitKat" ? Jan 23 02:32:15 because marketing Jan 23 02:32:22 Did they get payed by kitkat or was it just a funnyname.. Jan 23 02:33:04 there were rumors before that it should get called "key lime pie". was kinda confused when i read that they named it KitKat.. Jan 23 02:33:08 FabianH: mutually beneficial contract Jan 23 02:33:44 FabianH: the pre-release rumours are mostly a source of joy when the predictions are brutally murdered at google press conference Jan 23 02:34:19 I like "kitkat" Jan 23 02:34:23 i'm glad Fedora gave up on naming their releases, it gets lame after a while Jan 23 02:34:38 i'd rather just know iOS5 than fuckingfruitcake Jan 23 02:34:40 But why is there no androidified android that eats a piece of kitkat? :( Jan 23 02:35:14 FabianH: I guess a toy like androidify didn't get updated. Probably a thing on "idle" priority Jan 23 02:35:38 It did get updated in late december Jan 23 02:36:17 Maybe they just should release the source code.. It's just a great idea, Jan 23 02:36:27 That would help them promote the android love Jan 23 02:36:47 FabianH: yes, just saying that I doubt it has any priority at google. It's mostly a toy after all (although it's also responsible for my online avatar :D) Jan 23 02:37:27 Yes, it's responsible for almost all contact pictures in my phone :D Jan 23 02:38:25 Google has so many unnecessary (to lazy to google for right spelling), projects that they just should focus on this app, it's just fun as hell :D Jan 23 02:38:40 heh Jan 23 02:38:44 memes are so popular, people love this kind of stuff Jan 23 02:39:06 Hello Jan 23 02:39:08 Android team probably got a lot more to do, and I suspect Androidify was a fun side project Jan 23 02:39:13 Anyone with BLE 4.0 experience? Jan 23 02:39:43 the rest of google got some big projects going on... (also, I have to commend Google on OPSEC. Damn good at it) Jan 23 02:41:05 Am I the only one that thinks that the use of fragments is just to complex? Jan 23 02:41:18 no, you're not Jan 23 02:41:56 You have to do some repeating steps over and over, that's why i still use only activities, i hate unefficient programming, and writing 20 times the same code is just plain stupid for me.. Jan 23 02:42:29 In some situations, it might be usefull. Can help out when your making something for a tablet. Jan 23 02:42:38 well, we don't have macros, and the cult of broken inheritance coupled with lack of mixins did some serious damage to Java and programming in general Jan 23 02:42:39 I really hope the DEV team changes this ... Google shouldn't wonder that not many people are using fragments Jan 23 02:43:19 I dunno that fragments are necessarily tied in with macros or mixins, hm. Jan 23 02:43:37 @AndreYonadam, The idea is really amazing, but how it's implemented on the programmers side is just to weird ... Jan 23 02:43:51 kbs: no, but boilerplate repetition is :) Jan 23 02:43:59 (and unwieldy class structures) Jan 23 02:44:03 true enough. Jan 23 02:44:38 Guys do you have some advice on how to design a android app, that really is looking hot? Jan 23 02:44:49 FabianH: What do you mean? Jan 23 02:45:17 I know the design guidelines, but my apps just look like very weird.. the look like they are not even natvie apps.. Jan 23 02:45:45 grids are the secret to swiss-type design [which is where all these things are currently in...] Jan 23 02:45:50 When i try to design the ui, it's just like someone scrambled some eggs on the screens Jan 23 02:45:52 I don't know what to say. Try to use a lot of bright flat colors. Jan 23 02:46:04 Also try not to use a lot of images. Jan 23 02:46:11 no text either Jan 23 02:46:16 text isn't sexy Jan 23 02:46:18 haha Jan 23 02:46:20 cards ! Jan 23 02:46:26 Instead I just make background with like gradients and let android do the work. Jan 23 02:46:33 tiles, everyone loves tiles Jan 23 02:46:38 no.. i don't like cards.. they just look so odd.. Jan 23 02:46:55 what background do you choose? Jan 23 02:46:59 cards? Jan 23 02:47:04 I make my own. Jan 23 02:47:11 Or pattern Jan 23 02:47:11 i hope they go away or at least the design guidelines talk about why/when/where/how Jan 23 02:47:17 you have some example? Jan 23 02:47:21 umm Jan 23 02:47:22 yeah Jan 23 02:47:27 YES! Jan 23 02:47:34 the old graphic print masters are still the gold standard for how to do layouts. http://www.vignelli.com/canon.pdf Jan 23 02:48:09 there's actually a lot of overlap - and the swiss style is a fairly good fit with the design guidelines Jan 23 02:48:25 The problem is that i read 1001238123 articles about designing.. I can design websites quite nice, but as soon as it comes to android apps, i really don't know how.. Jan 23 02:48:53 start with grids :-) Jan 23 02:48:57 i'll read into swiss style.. you have some advice on font colors ? Jan 23 02:49:07 kbs: you stole my background Jan 23 02:49:07 haha Jan 23 02:49:08 damn Jan 23 02:49:09 In html i'm using grids Jan 23 02:49:32 someone call the police Jan 23 02:49:43 FabianH: why Jan 23 02:49:43 ha Jan 23 02:49:45 what kind of gradients do you use as a background? Jan 23 02:50:05 http://pastebin.com/ja97YPUb Jan 23 02:50:09 shapes Jan 23 02:50:18 and then add it as an xml Jan 23 02:50:20 for the bg Jan 23 02:50:24 idk Jan 23 02:50:27 you can probably use radial Jan 23 02:50:48 ok, thanks :D Jan 23 02:51:22 Are some android designers active on this channel, that work at goooogle? Jan 23 02:51:29 I guess Jan 23 02:51:43 What sensor would you guys use to find axis rotation? Jan 23 02:51:48 if yes, please notice: Get rid of the card layout, as you currently use it in play store.. Jan 23 02:52:24 I honestly don't know what a card layout is Jan 23 02:52:25 anyways Jan 23 02:52:26 FabianH cards seems to be part of their design language across several products, like Google+ Jan 23 02:52:41 i hate that thing too :) Jan 23 02:52:43 Idk if I can access the sensors directly from sensormanager Jan 23 02:52:53 i know, the implemented it, but it just looks so ... Jan 23 02:52:57 so old.. Jan 23 02:53:28 yeah, see what happens when you create superfluous crap :) Jan 23 02:53:54 stick to the basic principles, can't go wrong :) like kbs mentioned - grid systems, etc Jan 23 02:54:08 stupid little gimmicks don't age well Jan 23 02:54:27 what guys font size do you usally use for content text? 14 sp? regular font ? with what color on what background?? Jan 23 02:54:46 FabianH thats in the design guidelines Jan 23 02:54:48 that's the most difficult thing for me.. choosing a good textsize and textcolor to read.. Jan 23 02:54:56 Don't use fixed Jan 23 02:55:03 Can't you use an android one? Jan 23 02:55:12 14 sp is medium, but i see like many apps with different sizes Jan 23 02:55:33 Use styles Jan 23 02:55:38 TextAppearance Jan 23 02:55:39 haha Jan 23 02:55:40 Thats what I use Jan 23 02:55:42 Its way easier **** ENDING LOGGING AT Thu Jan 23 02:59:58 2014